Search is not available for this dataset
id
string | images
list | conversation
list | source
string |
|---|---|---|---|
train-3265-2-0
|
[
"nlvr2/train-3265-2-img0.png",
"nlvr2/train-3265-2-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\n<image> <image> Here is a statement describing these 2 images: The right image contains one adult red-orange chow standing in profile turned leftward, and the left image includes a fluffy young chow facing forward. Is it true or false? \nOptions:\n(A) True\n(B) False",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
llava_665k_merged_000000458617
|
[
"llava_665k_multi/coco/train2017/000000458617.jpg"
] |
[
{
"content": "<image> What is this photo about'?",
"role": "user"
},
{
"content": "The image features a dining table with a variety of sandwiches and sandwich ingredients. There is a plate filled with peanut butter and jelly sandwiches that have had their crusts cut off, making them similar to tea sandwiches. The sandwiches are situated in different locations on the table, some closer to the left, others at the center, and a few more toward the right.\n\nAdditionally, the table has various items including a couple of bottles, one placed near the top left corner while the other is situated towards the top right corner. A couple of spoons can be seen, with one near the top left and the other at the top middle part of the table. There is also a bowl located at the top left side and a cake on the middle-right side of the table.\n\nTwo knives are visible on the table, one positioned at the top center and the other at the top right corner. These tools were likely used to prepare the sandwiches and spread the ingredients.",
"role": "assistant"
}
] |
llava_665k_merged
|
train-6266-0-1
|
[
"nlvr2/train-6266-0-img0.png",
"nlvr2/train-6266-0-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\n<image> <image> Here is a statement describing these 2 images: In one image there is a single puppy sitting on the ground. Is it true or false? \nOptions:\n(A) False\n(B) True",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
train-4227-3-0
|
[
"nlvr2/train-4227-3-img0.png",
"nlvr2/train-4227-3-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nGiven these 2 images, <image> and <image>. Here is a statement describing them: A trifle is garnished with red fruit on white whipped cream. Is it true or false?\nOptions:\n(A) True\n(B) False",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
iconqa_train_choose_txt_kindergarten_C6_16398
|
[
"iconqa/iconqa_data/iconqa/train/choose_txt/63659/image.png"
] |
[
{
"content": "How many shapes are there?<image>\nA. 5\nB. 7\nC. 10\nD. 6\nE. 1",
"role": "user"
},
{
"content": "Answer: C",
"role": "assistant"
}
] |
iconqa
|
iconqa_train_fill_in_blank_grade2_A4_59481
|
[
"iconqa/iconqa_data/iconqa/train/fill_in_blank/49054/image.png"
] |
[
{
"content": "How many candies are there?<image>",
"role": "user"
},
{
"content": "Answer: 10",
"role": "assistant"
}
] |
iconqa
|
ImageCoDe-video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0-7
|
[
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img0.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img3.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img8.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img20.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img25.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img42.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img43.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img45.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img50.jpg",
"imagecode/image-sets/video-storytelling-videobirthday_3l0XYe5kjs4-shot57_0/img54.jpg"
] |
[
{
"content": "Given a detailed description, retrieve the target image among 10 minimally contrastive images\nDescription:\nThe girl is coming down the slide. Her hairline is not visible but her eyebrows and eyes are visible. \nA. <image>\nB. <image>\nC. <image>\nD. <image>\nE. <image>\nF. <image>\nG. <image>\nH. <image>\nI. <image>\nJ. <image>",
"role": "user"
},
{
"content": "Answer: H",
"role": "assistant"
}
] |
imagecode
|
1591373
|
[
"vg/VG_100K/2316069.jpg"
] |
[
{
"content": "Which of (0.71, 0.96) and (0.56, 0.87) is in the same object as (0.65, 0.88)? Options: (A) (0.56, 0.87), (B) (0.71, 0.96)\n<image>",
"role": "user"
},
{
"content": "(A) (0.56, 0.87)",
"role": "assistant"
}
] | null |
290158
|
[
"vg/VG_100K_2/2417033.jpg",
"vg/VG_100K/1591849.jpg",
"vg/VG_100K/2316519.jpg"
] |
[
{
"content": "<image> <image> <image>\nDetermine which image contains the most of kettle?",
"role": "user"
},
{
"content": "Image 0",
"role": "assistant"
}
] | null |
dream_sim_nights_1
|
[
"dreamsim/nights/distort/073/897_0.png",
"dreamsim/nights/ref/073/897.png",
"dreamsim/nights/distort/073/897_1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nIf image 2 is the reference image, which image of the other two is more similar to the reference image? <image> <image> <image>.\nOptions:\n(A) Image 1\n(B) Image 3",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
dreamsim
|
Lin-Chen/ShareGPT4V-6451-contrastive_caption_shuffle-24387-24390
|
[
"contrastive_caption/sharegpt4v/coco/train2017/000000121350.jpg",
"contrastive_caption/sharegpt4v/coco/train2017/000000121351.jpg",
"contrastive_caption/sharegpt4v/coco/train2017/000000121356.jpg"
] |
[
{
"content": "<image> <image> <image> What is visible in the following the second image",
"role": "user"
},
{
"content": "In the vast expanse of a sandy terrain, two majestic elephants command the scene. The elephant in the foreground, with its grayish skin reflecting the harsh sunlight, stands firmly with its trunk extended downwards, as if exploring the ground beneath. Its large tusks, a testament to its might, curve elegantly from its massive head. \n\nScattered around are rocks and debris, painting a picture of a rugged landscape that these magnificent creatures call home. A few steps behind, partially hidden from view, is the second elephant. It appears to be in motion, walking away from the camera, its back and partial profile visible.\n\nThe wrinkled skin of both elephants, a characteristic feature of their species, adds texture to the image. Despite the harsh environment, there's a sense of calmness that the image portrays - a snapshot of life in the wild.",
"role": "assistant"
},
{
"content": "What stands out in the the foremost image",
"role": "user"
},
{
"content": "This image captures a cozy living room scene, bathed in soft light. Dominating the center of the image is a vibrant yellow armchair, its color popping against the neutral tones of the room. The armchair, a symbol of comfort and relaxation, cradles a black cat curled up in sleep, adding a sense of tranquility to the scene.\n\nIn front of the armchair, a wooden coffee table holds a stack of magazines and a bowl of fruit, suggesting leisurely afternoons spent reading or enjoying a healthy snack. The table's rustic wooden texture contrasts nicely with the smooth fabric of the armchair.\n\nTo the right of the armchair, a red painting hangs on the white wall. Its bold color draws the eye and adds an artistic touch to the room. The painting's abstract design invites viewers to interpret its meaning.\n\nBehind the armchair, a cutout in the wall reveals a glimpse of the kitchen, connecting the two spaces and creating an open, airy feel. The kitchen's presence suggests that this living room is part of a larger living space where cooking and relaxation blend seamlessly together.\n\nThe floor is carpeted in beige, adding warmth and softness underfoot. The carpet's neutral color allows the yellow armchair and red painting to stand out as focal points in the room.\n\nOverall, this image portrays a comfortable and inviting living space where one can relax, read, enjoy a snack, or simply watch a beloved pet sleep.",
"role": "assistant"
},
{
"content": "Can you describe what's in the the secondary image",
"role": "user"
},
{
"content": "In the heart of a bustling city, a young girl radiates joy as she holds a delectable donut in her hands. The donut, a feast for the eyes and undoubtedly the taste buds, is generously adorned with chocolate chips and a layer of white frosting that contrasts beautifully with the dark chips. \n\nThe girl, dressed in a stylish black jacket over a pink shirt, is seated comfortably in a booth. Her sparkling eyes and wide smile reflect the anticipation of savoring the sweet treat she's holding. \n\nBehind her, a window reveals a slice of city life. A brick building stands tall, its red-brown hues adding warmth to the scene. Parked on the street adjacent to the building is a car, quietly witnessing the city's rhythm.\n\nEvery element in this image comes together to paint a vivid picture of a simple, joyful moment in a young girl's life in the city.",
"role": "assistant"
}
] |
contrastive_caption
|
coinstruct_125266
|
[
"coinstruct/data/israel-107946_21_05.bmp",
"coinstruct/data/0b66b5509328b89e357035bedc6d.jpg"
] |
[
{
"content": "The first image:<image>\nThe second image:<image>Compare the quality-related attributes and overall quality between the two images briefly. Your reference can be related to specific objects in the two images. Be short and precise.",
"role": "user"
},
{
"content": "The first image is a lower resolution image of a historic building, featuring a somewhat pixelated and blurry appearance, which reduces the level of detail and clarity. Specific objects such as stones and architectural features lack sharpness.\n\nThe second image is of higher resolution with clear and sharp details, particularly visible in the precise edges and reflective surfaces of the red objects. The lighting in the second image is also well captured, adding to the overall quality and aesthetic appeal.",
"role": "assistant"
}
] |
coinstruct
|
train-11802-3-1
|
[
"nlvr2/train-11802-3-img0.png",
"nlvr2/train-11802-3-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nGiven these 2 images, <image> and <image>. Here is a statement describing them: Multiple pairs of elephant tusks are visible. Is it true or false?\nOptions:\n(A) True\n(B) False",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
train-5115-0-0
|
[
"nlvr2/train-5115-0-img0.png",
"nlvr2/train-5115-0-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\n<image> <image> Here is a statement describing these 2 images: There is a spoon laying on the table near the bowl in one image. Is it true or false? \nOptions:\n(A) False\n(B) True",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
llava_665k_merged_000000436584
|
[
"llava_665k_multi/coco/train2017/000000436584.jpg"
] |
[
{
"content": "What type of photograph is the image of the outhouse? <image>",
"role": "user"
},
{
"content": "The image of the outhouse is a black and white photograph.",
"role": "assistant"
},
{
"content": "What is the condition of the outhouse in the image?",
"role": "user"
},
{
"content": "The outhouse in the image appears to be run-down, with a shabby door and roof.",
"role": "assistant"
},
{
"content": "What is visible inside the outhouse?",
"role": "user"
},
{
"content": "Inside the outhouse, we can see an open toilet. The flooring consists of tiles.",
"role": "assistant"
},
{
"content": "Can you briefly describe the features of an outhouse?",
"role": "user"
},
{
"content": "An outhouse is a small, simple, and typically single-occupancy structure that is used primarily as a toilet facility. Outhouses are often found in rural areas or locations where sewage connections and indoor plumbing are not available. The main features of an outhouse include:\n\n1. A small single room with a door for privacy.\n2. A toilet seat and hole, usually placed over a pit or a removable container for waste collection.\n3. Ventilation openings to allow air circulation and reduce odors.\n4. A simple roof and walls, usually made of wood or metal, to provide shelter from the elements.\n\nOuthouses may vary in appearance and construction materials, but they generally serve the same purpose – providing a basic, functional, and private space for individuals to use the toilet. In the image described, the outhouse appears to be run-down, with a shabby door and roof, signifying age, wear, or lack of maintenance.",
"role": "assistant"
},
{
"content": "What are some common reasons for using an outhouse instead of indoor plumbing?",
"role": "user"
},
{
"content": "There are several reasons why someone might use an outhouse instead of indoor plumbing. Some of the common reasons include:\n\n1. Limited access to infrastructure: In rural areas or regions where sewer connections and indoor plumbing are unavailable or difficult to access, an outhouse can provide a simple and cost-effective alternative.\n\n2. Temporary facilities: Outhouses are also used as temporary or portable facilities in places such as outdoor events, construction sites, or campgrounds, where more permanent solutions may be impractical or unnecessary.\n\n3. Cost and resource efficiency: Outhouses can be less expensive to build and maintain than septic systems or sewage treatments, especially in locations with limited water or sewer services.\n\n4. Environmental considerations: In some cases, outhouses can be a more environmentally friendly option by reducing water usage and avoiding the need for chemical treatments in the waste management process.\n\n5. Tradition: In certain cultures or historical settings, outhouses may be used simply because they have traditionally been the norm in those areas, and there may be a resistance to adopting more modern plumbing systems.\n\nIn short, the use of an outhouse can be an appropriate option for various practical, financial, and cultural reasons. However, it is also essential to ensure that outhouses are well-maintained and managed to minimize potential health and sanitation concerns.",
"role": "assistant"
}
] |
llava_665k_merged
|
chartqa_8041
|
[
"chartqa/two_col_40177.png"
] |
[
{
"content": "<image>How many Nobel Prize Laureates have been from the United States since 1901?",
"role": "user"
},
{
"content": "102",
"role": "assistant"
}
] |
chartqa
|
llava_665k_merged_142410373
|
[
"llava_665k_multi/ocr_vqa/images/142410373.jpg"
] |
[
{
"content": "<image> Who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Roald Dahl",
"role": "assistant"
},
{
"content": "What is the title of this book?",
"role": "user"
},
{
"content": "Matilda",
"role": "assistant"
},
{
"content": "What is the genre of this book?",
"role": "user"
},
{
"content": "Children's Books",
"role": "assistant"
},
{
"content": "Is this a kids book?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is this a sci-fi book?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
}
] |
llava_665k_merged
|
coinstruct_52079
|
[
"coinstruct/spaq_koniq/10602.jpg",
"coinstruct/spaq_koniq/00798.jpg"
] |
[
{
"content": "<image> <image> Which image has better quality, and why? Judge and explain briefly.",
"role": "user"
},
{
"content": "The first image has a better quality. This is because the image is clear and the details and texture of the main subject, the rainwater and fire hydrant, are presented well. The overall clarity is high, making it easy to distinguish the different elements in the image. In contrast, the second image has serious focusing issues, resulting in most of the content being blurred. The main subject, the goose, as well as the background sofa and textbook, are not distinguishable, making the overall clarity very low. Therefore, the quality of this image is poor compared to the first image.",
"role": "assistant"
}
] |
coinstruct
|
coinstruct_139640
|
[
"coinstruct/data/10541eb1c7b23c6ddfbac691225c620.jpg",
"coinstruct/data/historical-361800_09_05.bmp",
"coinstruct/data/16972b131db5e0747798113e882f4ee.jpg"
] |
[
{
"content": "<image> <image> <image> Compare the quality-related attributes and overall quality among the three images briefly. Your reference can be related to specific objects in the three images. Conclude with you final rank of quality among the images. Be consise and precise.",
"role": "user"
},
{
"content": "The three images differ significantly in quality mainly due to their resolution, clarity, and the presence of noise.\n\n1. The first image shows an overcast sky over a residential area at either dawn or dusk. The resolution and detail are low, making it difficult to discern finer features on the buildings. There is considerable noise, and the colors appear washed out, suggesting that the image may have been taken with a low-quality camera or in low lighting conditions.\n\n2. The second image is highly pixelated, and the object in question—a partial reflection in water—is almost indistinguishable due to the lack of detail and clarity. The colors are distorted, artifacts are present throughout, and it is difficult to interpret the content of the photo.\n\n3. The third image features a person in an indoor setting. The camera's motion blur, poor focus, and low-resolution result in a lack of sharpness and detail. The lighting conditions are poor, contributing to the overall low-quality appearance of the photo. However, despite the motion blur, this image has slightly better discernability of content compared to the second image.\n\nFinal ranking in terms of quality from highest to lowest would be: Image 1, Image 3, and Image 2. Image 1 is the most visually coherent despite its flaws, while Image 3 has slightly better content recognition than Image 2, which has the poorest quality of the three.",
"role": "assistant"
},
{
"content": "Which image has the lowest resolution?",
"role": "user"
},
{
"content": "Image 2",
"role": "assistant"
},
{
"content": "What common quality issue is present in all three images?\nA. Blur\nB. Overexposure\nC. Perfect quality\nAnswer with the option's letter from the given choices directly.",
"role": "user"
},
{
"content": "A.",
"role": "assistant"
},
{
"content": "Which image is taken outdoors?\nA. Image 1\nB. Image 2\nC. Image 3\nAnswer with the option's letter from the given choices directly.",
"role": "user"
},
{
"content": "B.",
"role": "assistant"
},
{
"content": "Is Image 3 blurry?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "How many images feature a natural environment?\nA. None\nB. One\nC. Two\nAnswer with the option's letter from the given choices directly.",
"role": "user"
},
{
"content": "B.",
"role": "assistant"
},
{
"content": "Which image has a person in it?\nA. Image 1\nB. Image 2\nC. Image 3\nAnswer with the option's letter from the given choices directly.",
"role": "user"
},
{
"content": "C.",
"role": "assistant"
},
{
"content": "Are there any images with good lighting conditions?\nA. Yes\nB. No\nAnswer with the option's letter from the given choices directly.",
"role": "user"
},
{
"content": "B.",
"role": "assistant"
},
{
"content": "Which image has the most visual noise?",
"role": "user"
},
{
"content": "Image 1",
"role": "assistant"
},
{
"content": "Is the person's face in Image 3 clear and in focus?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "Which image appears to have been taken during the daytime?\nA. Image 1\nB. Image 2\nC. Image 3\nAnswer with the option's letter from the given choices directly.",
"role": "user"
},
{
"content": "B.",
"role": "assistant"
}
] |
coinstruct
|
llava_665k_merged_2372528-2350144
|
[
"llava_665k_multi/gqa/images/2372528.jpg",
"llava_665k_multi/gqa/images/2350144.jpg"
] |
[
{
"content": "For the image on the left, what is the color of the train that is to the right of the other train? <image> <image>",
"role": "user"
},
{
"content": "Green",
"role": "assistant"
},
{
"content": "For the right image, does the white house look large?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the left image, is this a truck or a train?",
"role": "user"
},
{
"content": "Train",
"role": "assistant"
},
{
"content": "For the right image, what type of place is it?",
"role": "user"
},
{
"content": "Road",
"role": "assistant"
},
{
"content": "For the image on the left side, is the weather overcast?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image on the left-hand side, is the train that is to the left of the other train red and long?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image on the right side, what size is the black tire?",
"role": "user"
},
{
"content": "Large",
"role": "assistant"
},
{
"content": "For the image 2, is the truck to the right or to the left of the vehicle that is to the left of the van?",
"role": "user"
},
{
"content": "Left",
"role": "assistant"
},
{
"content": "For the image on the left-hand side, what color is the train to the left of the other train, red or gray?",
"role": "user"
},
{
"content": "Red",
"role": "assistant"
},
{
"content": "For the image 2, does the man that looks tall appear to be sitting?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image on the right side, is the bus to the left or to the right of the van?",
"role": "user"
},
{
"content": "Left",
"role": "assistant"
},
{
"content": "For the second image, on which side of the photo is the van?",
"role": "user"
},
{
"content": "Right",
"role": "assistant"
},
{
"content": "For the image on the left, how long is the red train?",
"role": "user"
},
{
"content": "Long",
"role": "assistant"
},
{
"content": "For the image 1, what color is the door?",
"role": "user"
},
{
"content": "Yellow",
"role": "assistant"
},
{
"content": "For the second image, what kind of vehicle is to the right of the truck?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Bus",
"role": "assistant"
},
{
"content": "For the image on the right, does the black tire look round and small?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image two, is the tall man to the left or to the right of the bus?",
"role": "user"
},
{
"content": "Right",
"role": "assistant"
},
{
"content": "For the second image, is the tall man to the right or to the left of the vehicle on the right side?",
"role": "user"
},
{
"content": "Left",
"role": "assistant"
},
{
"content": "For the image on the right hand side, is the tall man to the left of the bus in the picture?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image on the right side, on which side of the photo is the truck?",
"role": "user"
},
{
"content": "Left",
"role": "assistant"
},
{
"content": "For the image on the right side, what type of vehicle is to the left of the man that is not short?",
"role": "user"
},
{
"content": "Truck",
"role": "assistant"
},
{
"content": "For the image on the right hand side, is the tall man to the left or to the right of the vehicle that is on the left?",
"role": "user"
},
{
"content": "Right",
"role": "assistant"
},
{
"content": "For the image on the right, what vehicle is to the right of the bus?",
"role": "user"
},
{
"content": "Van",
"role": "assistant"
},
{
"content": "For the image on the right, is the white house on the left?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image on the right, is the house white and small?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
}
] |
llava_665k_merged
|
1333737
|
[
"vg/VG_100K/2330164.jpg"
] |
[
{
"content": "<image>\nDetermine how are objects at (0.58, 0.32, 0.97, 0.7) and (0.24, 0.42, 0.36, 0.57) related?\nSelections: (A) skiing in, (B) growing next to, (C) standing by, (D) to the right of",
"role": "user"
},
{
"content": "(D) to the right of",
"role": "assistant"
}
] | null |
Lin-Chen/ShareGPT4V-18079-contrastive_caption_shuffle-79442-79448
|
[
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000136495.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000136495.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000136495.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000136498.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000136498.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000136498.jpg"
] |
[
{
"content": "What elements are present in the the opening image <image> <image> <image> <image> <image> <image>",
"role": "user"
},
{
"content": "The image captures a charming scene of various objects arranged on a wooden surface. Dominating the left side of the frame is a gray elephant figurine, standing still and appearing to be made of a smooth, solid material. \n\nMoving towards the right, we encounter a collection of fruit. A green pear, a red apple, and a yellow banana are all present, their vibrant colors contrasting beautifully with the wooden background. They are not in any specific order but are loosely grouped together.\n\nFurther to the right, an orange coil rests. Its bright color and unique shape add an element of intrigue to the scene.\n\nIn the lower right corner of the image, a quarter is visible. It's not interacting with the other objects but provides a helpful reference for scale, allowing us to understand the relative sizes of the other items in the scene.\n\nThe objects are all inanimate and stationary, each contributing to the overall composition of the image. The relative positions of the objects and their distinct colors create a visually engaging scene. The image does not contain any discernible text. The overall arrangement seems casual, with no apparent pattern or alignment among the objects. The image is rich in detail and offers a snapshot of a simple yet captivating arrangement of everyday objects.",
"role": "assistant"
},
{
"content": "What can you see in the the image directly after the first",
"role": "user"
},
{
"content": "The image captures a charming scene on a wooden surface. At the heart of the scene is a US quarter, around which a collection of small objects are arranged in a loose semi-circle. A gray elephant figurine, made of what appears to be clay or plasticine, stands facing the quarter. The elephant's companions, a green pear, a red apple, a yellow banana, and an orange coil, are all facing away from the quarter. Each object, with its distinct color and shape, adds a unique element to the scene. The arrangement of these objects suggests a playful and creative atmosphere.",
"role": "assistant"
},
{
"content": "What are the key components of the the image following the fifth image",
"role": "user"
},
{
"content": "In the image, a cactus plant is captured in a close-up shot. The cactus, exhibiting a pale green color, is covered with numerous small white spines. These spines are arranged in a circular pattern, particularly noticeable around the top of the plant. The cactus displays a unique curvature, with the top part bending towards the right side of the frame. This creates an interesting contrast with the clear blue sky that forms the backdrop of the image. The image does not contain any discernible text. The relative position of the cactus and the sky suggests that the photo was likely taken from a low angle, looking upwards. The cactus, being the only object in the image, stands out prominently against the sky.",
"role": "assistant"
},
{
"content": "What are the key components of the image 4",
"role": "user"
},
{
"content": "The image captures a single cactus flower bud in the foreground, poised against a backdrop of a clear blue sky. The bud, covered in small white spines, is perched at the end of a long green stem that curves gracefully from the bottom left corner of the image to the top right corner. The spines on the bud are arranged in a circular pattern, with the ones in the center noticeably longer than those on the edges. The stark contrast between the green stem and the blue sky, along with the intricate pattern of the spines, creates a captivating visual. The image does not contain any discernible text or additional objects. The relative position of the bud at the top right corner gives the image a sense of depth and perspective. The bud, being the only object in the image, stands out prominently. The image is a close-up shot, allowing for the detailed observation of the bud and its spines. The bud appears to be in a state of growth, hinting at the potential of a blooming flower in the near future. The image, in its simplicity and detail, provides a glimpse into the beauty of nature's creations.",
"role": "assistant"
},
{
"content": "What is in the the third image",
"role": "user"
},
{
"content": "The image captures a charming scene of various objects casually arranged on a wooden surface. Dominating the left side of the frame is a gray elephant figurine, standing still and appearing to be made of a smooth, solid material. \n\nMoving towards the right, we encounter a vibrant display of fruit. A green pear, a red apple, and a yellow banana are all present, their bright colors contrasting beautifully with the wood beneath. They are not in any specific order but are loosely grouped together.\n\nFurther to the right, an orange coil rests, its spiraling form adding a touch of whimsy to the scene. \n\nIn the lower right corner of the image, a quarter lies flat on the surface, its presence providing a helpful reference for the scale of the other objects. \n\nThe overall arrangement of the objects seems casual, with no apparent pattern or order. Yet, each object holds its own space, contributing to a harmonious composition that is both simple and intriguing.",
"role": "assistant"
},
{
"content": "What stands out in the the last image",
"role": "user"
},
{
"content": "The image captures a close-up view of a cactus plant, bathed in the light of a clear day. The cactus, exhibiting a pale green hue, stands out against the backdrop of a clear blue sky. The surface of the cactus is densely covered with small white spines, giving it a textured appearance. These spines are arranged in a circular pattern, particularly noticeable around the top of the cactus, adding a sense of symmetry to the image. The cactus, with its intricate pattern of spines, commands the viewer's attention, making it the focal point of the image. The clear blue sky in the background provides a stark contrast, further emphasizing the cactus. There are no discernible texts or other objects in the image. The relative position of the cactus against the sky suggests that the photo was taken from a low angle, looking upwards. The image does not contain any action, but rather captures a still moment in time, focusing on the detailed structure of the cactus.",
"role": "assistant"
}
] |
contrastive_caption
|
llava_665k_merged_1448874335
|
[
"llava_665k_multi/ocr_vqa/images/1448874335.jpg"
] |
[
{
"content": "Who wrote this book?\nAnswer the question using a single word or phrase. <image>",
"role": "user"
},
{
"content": "Leslie Harper",
"role": "assistant"
},
{
"content": "What is the title of this book?",
"role": "user"
},
{
"content": "What Are Checks and Balances? (Civics Q&A (Powerkids))",
"role": "assistant"
},
{
"content": "What type of book is this?",
"role": "user"
},
{
"content": "Children's Books",
"role": "assistant"
},
{
"content": "Is this book related to Children's Books?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is this book related to Gay & Lesbian?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
}
] |
llava_665k_merged
|
1951037
|
[
"vg/VG_100K/1159450.jpg"
] |
[
{
"content": "Identify which object is positioned closer to the viewpoint, in skis and mountain? Options: (A) skis, (B) mountain\n<image>",
"role": "user"
},
{
"content": "(A) skis",
"role": "assistant"
}
] | null |
dvqa_reasoning_question_119719502_119719503_119719504_119719505_119719506_119719507_119719510_119719511_119719512_119719513_119719517_119719518
|
[
"dvqa/bar_train_00197195.png"
] |
[
{
"content": "<image>How many objects are preferred by less than 1 people in at least one category?",
"role": "user"
},
{
"content": "zero",
"role": "assistant"
},
{
"content": "Which object is the most preferred in any category?",
"role": "user"
},
{
"content": "fuel",
"role": "assistant"
},
{
"content": "How many people like the most preferred object in the whole chart?",
"role": "user"
},
{
"content": "9",
"role": "assistant"
},
{
"content": "Which object is preferred by the least number of people summed across all the categories?",
"role": "user"
},
{
"content": "finger",
"role": "assistant"
},
{
"content": "Which object is preferred by the most number of people summed across all the categories?",
"role": "user"
},
{
"content": "fuel",
"role": "assistant"
},
{
"content": "How many total people preferred the object fury across all the categories?",
"role": "user"
},
{
"content": "9",
"role": "assistant"
},
{
"content": "Is the object gun in the category angle preferred by less people than the object finger in the category blood?",
"role": "user"
},
{
"content": "no",
"role": "assistant"
},
{
"content": "What category does the steelblue color represent?",
"role": "user"
},
{
"content": "blood",
"role": "assistant"
},
{
"content": "How many people prefer the object fury in the category angle?",
"role": "user"
},
{
"content": "3",
"role": "assistant"
},
{
"content": "What is the label of the third group of bars from the left?",
"role": "user"
},
{
"content": "prize",
"role": "assistant"
},
{
"content": "What is the label of the second bar from the left in each group?",
"role": "user"
},
{
"content": "angle",
"role": "assistant"
},
{
"content": "Is each bar a single solid color without patterns?",
"role": "user"
},
{
"content": "no",
"role": "assistant"
},
{
"content": "How many groups of bars are there?",
"role": "user"
},
{
"content": "seven",
"role": "assistant"
}
] |
dvqa
|
train-300-3-0
|
[
"nlvr2/train-300-3-img0.png",
"nlvr2/train-300-3-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nGiven these 2 images, <image> and <image>. Here is a statement describing them: In one of the images there is a bookstore without any shoppers. Is it true or false?\nOptions:\n(A) False\n(B) True",
"role": "user"
},
{
"content": "B",
"role": "assistant"
}
] |
nlvr2
|
2069657
|
[
"vg/VG_100K/2360555.jpg"
] |
[
{
"content": "Tell me in the list of hands and bookshelves, which object is positioned farther away to the viewpoint?\n<image>",
"role": "user"
},
{
"content": "bookshelves",
"role": "assistant"
}
] | null |
iconqa_train_choose_img_grade1_R2_19765
|
[
"iconqa/iconqa_data/iconqa/train/choose_img/45124/image.png",
"iconqa/iconqa_data/iconqa/train/choose_img/45124/choice_0.png",
"iconqa/iconqa_data/iconqa/train/choose_img/45124/choice_1.png"
] |
[
{
"content": "What shape comes next?<image>\nA. <image>\nB. <image>",
"role": "user"
},
{
"content": "Answer: A",
"role": "assistant"
}
] |
iconqa
|
llava_665k_merged_000000050305-000000050305-000000577488
|
[
"llava_665k_multi/coco/train2017/000000050305.jpg",
"llava_665k_multi/coco/train2017/000000050305.jpg",
"llava_665k_multi/coco/train2017/000000577488.jpg"
] |
[
{
"content": "For the image 1, please provide the bounding box coordinate of the region this sentence describes: a boy with glasses , wearing a white shirt with a tan messener bag. <image> <image> <image>",
"role": "user"
},
{
"content": "[0.54, 0.42, 0.73, 0.96]",
"role": "assistant"
},
{
"content": "For the image on the left, please provide a short description for this region: [0.54, 0.42, 0.73, 0.96].",
"role": "user"
},
{
"content": "Right guy.",
"role": "assistant"
},
{
"content": "For the left image, please provide a short description for this region: [0.32, 0.41, 0.45, 0.97].",
"role": "user"
},
{
"content": "Boy on left.",
"role": "assistant"
},
{
"content": "For the image 1, please provide the bounding box coordinate of the region this sentence describes: boy holding umbrella.",
"role": "user"
},
{
"content": "[0.42, 0.4, 0.59, 0.96]",
"role": "assistant"
},
{
"content": "For the image on the left, please provide the bounding box coordinate of the region this sentence describes: shortest person white shirt on end.",
"role": "user"
},
{
"content": "[0.54, 0.42, 0.73, 0.96]",
"role": "assistant"
},
{
"content": "For the image on the middle, please provide a short description for this region: [0.32, 0.41, 0.45, 0.97].",
"role": "user"
},
{
"content": "Person on the left.",
"role": "assistant"
},
{
"content": "For the image on the left-hand side, please provide the bounding box coordinate of the region this sentence describes: guy holding umbrella.",
"role": "user"
},
{
"content": "[0.42, 0.4, 0.59, 0.96]",
"role": "assistant"
},
{
"content": "For the image three, please provide a short description for this region: [0.59, 0.44, 0.84, 0.64].",
"role": "user"
},
{
"content": "Two horses tied up to the front of a cart.",
"role": "assistant"
},
{
"content": "For the final image, please provide a short description for this region: [0.59, 0.44, 0.84, 0.64].",
"role": "user"
},
{
"content": "Two black horses pulling a wagon.",
"role": "assistant"
},
{
"content": "For the left image, please provide a short description for this region: [0.42, 0.4, 0.59, 0.96].",
"role": "user"
},
{
"content": "Guy in middle.",
"role": "assistant"
}
] |
llava_665k_merged
|
iconqa_train_choose_txt_prek_B3_13325
|
[
"iconqa/iconqa_data/iconqa/train/choose_txt/84306/image.png"
] |
[
{
"content": "How many cars are there?<image>\nA. 2\nB. 4\nC. 5\nD. 1\nE. 3",
"role": "user"
},
{
"content": "Answer: C",
"role": "assistant"
}
] |
iconqa
|
866507
|
[
"vg/VG_100K/2324069.jpg",
"vg/VG_100K/2373830.jpg",
"vg/VG_100K/2342099.jpg"
] |
[
{
"content": "<image> <image> <image>\nTell me what differences can be observed in the relation between blanket and grass?\nChoose from: A. blanket is on grass in Image 0, to the right of grass in Image 1, to the left of grass in Image 2., B. blanket is on grass in Image 0, to the left of grass in Image 1, to the right of grass in Image 2., C. blanket is to the right of grass in Image 0, on grass in Image 1, to the left of grass in Image 2., D. blanket is to the left of grass in Image 0, to the right of grass in Image 1, on grass in Image 2.",
"role": "user"
},
{
"content": "B. blanket is on grass in Image 0, to the left of grass in Image 1, to the right of grass in Image 2.",
"role": "assistant"
}
] | null |
251119
|
[
"vg/VG_100K_2/2402541.jpg"
] |
[
{
"content": "Determine which object is located on the far left, out of curtain, blinds, and door?\n<image>",
"role": "user"
},
{
"content": "blinds",
"role": "assistant"
}
] | null |
dream_sim_nights_17845
|
[
"dreamsim/nights/ref/089/256.png",
"dreamsim/nights/distort/089/256_0.png",
"dreamsim/nights/distort/089/256_1.png"
] |
[
{
"content": "Answer the following question:\nHere are there images: <image> <image> <image>. If image 1 is the reference image, which image of the other two is more similar to the reference image?\n",
"role": "user"
},
{
"content": "Image 2",
"role": "assistant"
}
] |
dreamsim
|
llava_665k_merged_000000253109-000000354766-000000526739-000000046106
|
[
"llava_665k_multi/coco/train2017/000000253109.jpg",
"llava_665k_multi/coco/train2017/000000354766.jpg",
"llava_665k_multi/coco/train2017/000000526739.jpg",
"llava_665k_multi/coco/train2017/000000046106.jpg"
] |
[
{
"content": "For the image 4, why is this woman wearing a coat?\nAnswer the question using a single word or phrase. <image> <image> <image> <image>",
"role": "user"
},
{
"content": "Cold",
"role": "assistant"
},
{
"content": "For the image 1, what are the traffic rules when there are four signs like these at each road of crossroads?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "4 way stop",
"role": "assistant"
},
{
"content": "For the image 3, what president of the usa is this toy named after?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Roosevelt",
"role": "assistant"
},
{
"content": "For the image 2, the orange flowers are called what?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Lily",
"role": "assistant"
}
] |
llava_665k_merged
|
Lin-Chen/ShareGPT4V-18310-contrastive_caption_select-80316-80324
|
[
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138253.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138253.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138253.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138253.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138253.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138257.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138257.jpg",
"contrastive_caption/sharegpt4v/llava/llava_pretrain/images/00013/000138257.jpg"
] |
[
{
"content": "Which image is the caption describing?\nIn the image, a lively scene unfolds on a rooftop patio against the backdrop of a bustling city skyline punctuated by tall buildings and construction cranes. The patio is furnished with a table, which is the focal point of the gathering. The table, equipped with a fire pit in its center, is encircled by black chairs. \n\nA group of people are seen congregating around this table, engaged in what appears to be a convivial gathering. Their positions around the table suggest a sense of camaraderie and shared enjoyment. \n\nOn the left side of the image, a man is holding a camera, capturing the moment. His actions indicate that he is filming the group, adding a layer of dynamism to the scene. The precise positioning of the objects and people, along with their actions, paints a vivid picture of a social gathering in an urban setting. <image> <image> <image> <image> <image> <image> <image> <image>",
"role": "user"
},
{
"content": "Image 7",
"role": "assistant"
},
{
"content": "Which image is the caption related to?\nThe image presents a charming scene of a miniature winter village, meticulously crafted into an advent calendar. The calendar, predominantly white, is composed of 24 individual drawers, each marked with a number from 1 to 24. These drawers form the base of the village, serving as a foundation for the structures above.\n\nAtop the drawers, small houses and trees are arranged, creating a picturesque village landscape. The houses, also white, are adorned with gold roofs, adding a touch of elegance to the scene. The trees, in contrast, are a vibrant green, providing a pop of color against the otherwise monochromatic palette.\n\nThe entire calendar is set against a gray background, which further accentuates the details of the village. Despite its simplicity, the calendar is a beautiful representation of a winter village, capturing the essence of the holiday season in a unique and creative way. The precise arrangement of the houses and trees, along with the numbered drawers, adds a sense of order and anticipation, characteristic of advent calendars. \n\nHowever, the image does not contain any discernible text or actions, and the relative positions of the objects remain constant, as they are part of a single, stationary piece. The image is a testament to the beauty of simplicity and the joy of anticipation that advent calendars bring during the holiday season.",
"role": "user"
},
{
"content": "The first image",
"role": "assistant"
},
{
"content": "Which image is the caption referring to?\nThe image captures a lively scene on a rooftop patio in the midst of urban surroundings. A group of people are gathered around a rectangular fire pit made of concrete, its warm glow contrasting with the cool tones of the cityscape. The individuals appear engaged in conversation, their attention drawn to each other rather than the impressive view that surrounds them. One person is holding a camera, perhaps capturing the moment for posterity.\n\nThe patio furniture, modern in style and black in color, complements the contemporary vibe of the setting. A table set with glasses and a candle suggests a casual social gathering. The rooftop is nestled among tall buildings, and a construction crane can be seen in the background, indicating ongoing development in the area. The image exudes a sense of camaraderie and urban living, encapsulating a moment of relaxation amidst the hustle and bustle of city life.",
"role": "user"
},
{
"content": "Image 6",
"role": "assistant"
},
{
"content": "Which image is the caption associated with?\nThe image presents a charming scene of a miniature winter village, meticulously crafted as an advent calendar. The calendar, predominantly white, stands out against a gray background. It's designed with 24 numbered drawers, each representing a day leading up to Christmas. \n\nThe top of the calendar is adorned with a variety of structures and trees, creating the illusion of a small, festive village. The houses, varying in size and design, are primarily white, with some featuring green roofs. The trees, scattered amongst the houses, are painted in shades of white and green, with a few boasting a touch of gold, adding a hint of sparkle to the scene. \n\nDespite its compact size, the calendar captures the essence of a bustling village preparing for the holiday season. The positioning of the houses and trees suggests a sense of community, with each element carefully placed in relation to the others. The numbered drawers, neatly arranged at the base of the calendar, invite anticipation and excitement for the days leading up to Christmas. \n\nOverall, the image portrays a delightful advent calendar, designed as a miniature village, ready to bring joy and holiday spirit to any home.",
"role": "user"
},
{
"content": "The second image",
"role": "assistant"
},
{
"content": "Which image is the caption about?\nIn the heart of a bustling city, a group of people have gathered on a rooftop patio for a convivial evening. The centerpiece of their gathering is a table, adorned with a fire pit that casts a warm glow on the faces of the attendees. The table is encircled by black chairs, inviting the guests to sit and enjoy the company.\n\nOn the left side of the photo, a man is capturing the moment with his camera, perhaps to immortalize the joyous occasion. His focus is intense, his fingers deftly adjusting the camera settings to get the perfect shot.\n\nOn the right side of the photo, another man is caught in a moment of laughter, his face lit up with mirth. His joy is infectious, adding to the overall cheerfulness of the gathering.\n\nA woman is seated at the table, her back to the camera, engrossed in the conversation around her. Her posture suggests she is relaxed and comfortable in the company of her friends.\n\nThe patio offers a stunning view of the city skyline, the buildings standing tall against the sky, their lights twinkling like stars. It's a beautiful backdrop to an evening of camaraderie and laughter. The city seems to be a silent observer of the festivities, its lights reflecting the warmth of the gathering on the rooftop. \n\nThis image captures a moment of joy, friendship, and warmth, set against the backdrop of a vibrant city. It's a snapshot of life, filled with laughter, conversations, and shared experiences.",
"role": "user"
},
{
"content": "The eighth image",
"role": "assistant"
},
{
"content": "Which image do you think the caption belongs to?\nThe image presents a charming advent calendar, designed to resemble a quaint village on a winter's day. The calendar is predominantly white, creating a snowy effect, and is set against a contrasting gray background. \n\nThe calendar is meticulously crafted into the shape of a village, complete with houses and trees. The houses, varying in sizes and shapes, are neatly arranged, creating a harmonious balance. They are adorned in shades of white and gold, while the trees are a refreshing green, adding a touch of color to the otherwise monochrome palette.\n\nThe unique feature of this advent calendar is its 24 numbered drawers, each representing a day leading up to Christmas. These drawers are cleverly integrated into the design, maintaining the aesthetic appeal while serving a functional purpose.\n\nThe overall design and attention to detail in the calendar evoke a sense of anticipation and excitement, characteristic of the advent season.",
"role": "user"
},
{
"content": "The fifth image",
"role": "assistant"
},
{
"content": "Which image is the caption linked to?\nThe image showcases an advent calendar, meticulously crafted to resemble a quaint little village. The calendar, predominantly white, is composed of 24 individually numbered drawers, each representing a day leading up to Christmas. The top of the calendar is adorned with miniature houses and trees, painted in hues of white, gold, and green, adding a festive touch to the overall design. The entire setup is placed against a contrasting gray background, making the calendar stand out prominently. The relative positions of the houses and trees on top create a picturesque scene of a serene winter village. The calendar, with its intricate design and attention to detail, serves as a charming countdown to the holiday season.",
"role": "user"
},
{
"content": "The third image",
"role": "assistant"
},
{
"content": "Which image is the caption connected to?\nThe image presents a charming advent calendar, designed to resemble a quaint village on a winter's day. The calendar is predominantly white, creating a snowy landscape effect. It features 24 numbered drawers, each representing a day leading up to Christmas. \n\nThe calendar is meticulously crafted in the shape of a village, complete with small houses and trees. The houses, varying in size and design, are neatly arranged in rows. Some of them are adorned with gold accents, adding a festive touch to the scene. \n\nThe trees, painted in shades of green and gold, are interspersed among the houses, enhancing the village's charm. The entire setup is placed against a gray background, which further accentuates the white and gold elements of the calendar.\n\nThe relative positions of the houses and trees create a sense of depth, making the village appear three-dimensional. Despite the absence of any discernible action, the calendar exudes a sense of anticipation, characteristic of the advent season. There are no texts or inscriptions visible in the image. The overall composition is balanced, with each element contributing to the festive theme. \n\nThis detailed description is based on the visible content of the image and does not include any speculative or imaginary elements.",
"role": "user"
},
{
"content": "The final image",
"role": "assistant"
}
] |
contrastive_caption
|
train-8709-3-0
|
[
"nlvr2/train-8709-3-img0.png",
"nlvr2/train-8709-3-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\n<image> <image> Here is a statement describing these 2 images: One image shows a single squarish bottle to the right of its upright case. Is it true or false? \nOptions:\n(A) True\n(B) False",
"role": "user"
},
{
"content": "B",
"role": "assistant"
}
] |
nlvr2
|
726619
|
[
"vg/VG_100K_2/3201.jpg"
] |
[
{
"content": "Tell me what is the number of tall street light?\n<image>",
"role": "user"
},
{
"content": "1",
"role": "assistant"
}
] | null |
llava_665k_merged_1580017428
|
[
"llava_665k_multi/ocr_vqa/images/1580017428.jpg"
] |
[
{
"content": "<image> Who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "International Code Council",
"role": "assistant"
},
{
"content": "What is the title of this book?",
"role": "user"
},
{
"content": "2009 International Energy Conservation Code: Softcover Version (International Code Council Series)",
"role": "assistant"
},
{
"content": "What is the genre of this book?",
"role": "user"
},
{
"content": "Crafts, Hobbies & Home",
"role": "assistant"
},
{
"content": "Is this a crafts or hobbies related book?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is this a comedy book?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
}
] |
llava_665k_merged
|
dvqa_reasoning_question_13721402_13721403_13721404_13721405_13721406_13721408_13721409_13721410_13721411_13721412_13721414
|
[
"dvqa/bar_train_00037214.png"
] |
[
{
"content": "How many groups of bars contain at least one bar with value smaller than 0?<image>",
"role": "user"
},
{
"content": "zero",
"role": "assistant"
},
{
"content": "Which group of bars contains the largest valued individual bar in the whole chart?",
"role": "user"
},
{
"content": "birth",
"role": "assistant"
},
{
"content": "What is the value of the largest individual bar in the whole chart?",
"role": "user"
},
{
"content": "9",
"role": "assistant"
},
{
"content": "Which group has the smallest summed value?",
"role": "user"
},
{
"content": "focus",
"role": "assistant"
},
{
"content": "Which group has the largest summed value?",
"role": "user"
},
{
"content": "birth",
"role": "assistant"
},
{
"content": "Is the value of movie in heat smaller than the value of birth in field?",
"role": "user"
},
{
"content": "yes",
"role": "assistant"
},
{
"content": "Are the values in the chart presented in a percentage scale?",
"role": "user"
},
{
"content": "no",
"role": "assistant"
},
{
"content": "What element does the forestgreen color represent?",
"role": "user"
},
{
"content": "field",
"role": "assistant"
},
{
"content": "What is the value of field in police?",
"role": "user"
},
{
"content": "4",
"role": "assistant"
},
{
"content": "What is the label of the first group of bars from the left?",
"role": "user"
},
{
"content": "movie",
"role": "assistant"
},
{
"content": "What is the label of the first bar from the left in each group?",
"role": "user"
},
{
"content": "stand",
"role": "assistant"
},
{
"content": "Are the bars horizontal?",
"role": "user"
},
{
"content": "no",
"role": "assistant"
}
] |
dvqa
|
llava_665k_merged_1939130387-873718968-1930556098-1441957677
|
[
"llava_665k_multi/ocr_vqa/images/1939130387.jpg",
"llava_665k_multi/ocr_vqa/images/873718968.jpg",
"llava_665k_multi/ocr_vqa/images/1930556098.jpg",
"llava_665k_multi/ocr_vqa/images/1441957677.jpg"
] |
[
{
"content": "<image> <image> <image> <image> For the image 2, is this book related to Romance?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 2, what is the title of this book?",
"role": "user"
},
{
"content": "Chemical Safety in the Laboratory",
"role": "assistant"
},
{
"content": "For the image 1, is this book related to Comics & Graphic Novels?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image 2, is this book related to Science & Math?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image 1, who wrote this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Fumi Yoshinaga",
"role": "assistant"
},
{
"content": "For the image 4, is this a kids book?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 3, what is the title of this book?",
"role": "user"
},
{
"content": "Dealing With Difficult Parents And With Parents in Difficult Situations",
"role": "assistant"
},
{
"content": "For the image 3, what type of book is this?",
"role": "user"
},
{
"content": "Education & Teaching",
"role": "assistant"
},
{
"content": "For the image 2, who wrote this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Stephen K. Hall",
"role": "assistant"
},
{
"content": "For the image 3, is this book related to Health, Fitness & Dieting?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 4, what is the title of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Advanced Flip Chip Packaging",
"role": "assistant"
},
{
"content": "For the image 4, what is the genre of this book?",
"role": "user"
},
{
"content": "Science & Math",
"role": "assistant"
},
{
"content": "For the image 3, is this book related to Education & Teaching?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image 1, what is the title of this book?",
"role": "user"
},
{
"content": "What Did You Eat Yesterday?, Volume 1",
"role": "assistant"
},
{
"content": "For the image 3, who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Todd Whitaker",
"role": "assistant"
},
{
"content": "For the image 1, is this book related to Computers & Technology?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 1, what type of book is this?",
"role": "user"
},
{
"content": "Comics & Graphic Novels",
"role": "assistant"
},
{
"content": "For the image 2, what type of book is this?",
"role": "user"
},
{
"content": "Science & Math",
"role": "assistant"
}
] |
llava_665k_merged
|
docvqa_ktxw0227_question_63300
|
[
"docvqa/ktxw0227_19.png"
] |
[
{
"content": "What is the \"Subject of the procedure\" ?<image>",
"role": "user"
},
{
"content": "Adult/Child Abuse and/or Neglect",
"role": "assistant"
}
] |
docvqa
|
train-11753-0-0
|
[
"nlvr2/train-11753-0-img0.png",
"nlvr2/train-11753-0-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nGiven these 2 images, <image> and <image>. Here is a statement describing them: The image on the left contains at least five wolves that are howling. Is it true or false?\nOptions:\n(A) True\n(B) False",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
docvqa_qpjf0226_question_771
|
[
"docvqa/qpjf0226_4.png"
] |
[
{
"content": "What is the specialization of international litigation services?<image>",
"role": "user"
},
{
"content": "international cases and foreign language document processing",
"role": "assistant"
}
] |
docvqa
|
train-5505-1-0
|
[
"nlvr2/train-5505-1-img0.png",
"nlvr2/train-5505-1-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nGiven these 2 images, <image> and <image>. Here is a statement describing them: One image shows a single reclining white puppy with at least one front paw forward, and the other image shows a white dog reclining with a 'real' sleeping hooved animal. Is it true or false?\nOptions:\n(A) False\n(B) True",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
coinstruct_59973
|
[
"coinstruct/spaq_koniq/06336.jpg",
"coinstruct/spaq_koniq/3230415578.jpg"
] |
[
{
"content": "<image> <image> Compare the quality between the first image and the second image.",
"role": "user"
},
{
"content": "The first image has poor quality due to its low clarity, focusing issues, noise, and monotone colors. The main subjects appear blurry, and most of the texture details are lost, making it difficult to make out the details of the sofa and table.\n\nThe second image has better quality than the first one, as it has good lighting and is focused on the upper part of the subject. The texture details are clear in the upper part, but the lower part and the features of Beijing are blurry, and the composition is incomplete. Although the image is not perfect, it is still better than the first one due to its better lighting, clearer texture details, and more vibrant colors. However, it is important to note",
"role": "assistant"
}
] |
coinstruct
|
llava_665k_merged_000000045360
|
[
"llava_665k_multi/coco/train2017/000000045360.jpg"
] |
[
{
"content": "<image> What potential issues might the zebra face in its current environment?",
"role": "user"
},
{
"content": "In the image, the zebra is grazing on a grass-covered field, which appears to be mainly dirt or muddy. This might pose several potential issues for the zebra. Firstly, limited grass availability due to the muddy conditions could make it challenging for the zebra to find sufficient food sources. Secondly, the muddy environment can lead to a variety of health issues, such as hoof problems or infections, if the zebra spends prolonged periods in the wet and muddy conditions. Additionally, the sparse vegetation might make the zebra more visible and susceptible to predators, increasing its vulnerability to threats. Thus, the zebra has to adapt and possibly move to other areas with greater vegetation and better conditions to overcome these challenges.",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_819562017
|
[
"llava_665k_multi/ocr_vqa/images/819562017.jpg"
] |
[
{
"content": "<image> Who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Daniel Walker Howe",
"role": "assistant"
},
{
"content": "What is the title of this book?",
"role": "user"
},
{
"content": "The Unitarian Conscience: Harvard Moral Philosophy, 1805-1861",
"role": "assistant"
},
{
"content": "What is the genre of this book?",
"role": "user"
},
{
"content": "Religion & Spirituality",
"role": "assistant"
},
{
"content": "Is this book related to Religion & Spirituality?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is this book related to Test Preparation?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_000000139249-000000296348-000000448851
|
[
"llava_665k_multi/coco/train2017/000000139249.jpg",
"llava_665k_multi/coco/train2017/000000296348.jpg",
"llava_665k_multi/coco/train2017/000000448851.jpg"
] |
[
{
"content": "For the image on the middle side, what is an alternative way to serve these hot dogs for a more health-conscious option? <image> <image> <image>",
"role": "user"
},
{
"content": "An alternative method to serve these hot dogs for a more health-conscious option would be to remove the bacon wrapping and consider using turkey, chicken, or plant-based hot dogs instead of traditional pork or beef franks. By eliminating the bacon and opting for a leaner or plant-based protein, you reduce saturated fat and calories in the dish, making it a more suitable choice for health-conscious individuals. Additionally, you can serve the hot dogs with whole wheat or whole grain buns for added fiber and nutrients, and include a variety of nutrient-rich vegetables or salads as side dishes to provide a healthier and more balanced meal.",
"role": "assistant"
},
{
"content": "For the third image, what challenges might the presence of these cats cause in this scene?",
"role": "user"
},
{
"content": "The presence of two cats lounging on the table covered with various objects, like a cup and a bottle, could pose challenges related to cleanliness and the potential for accidents. Cats are known to shed fur and could accidentally knock over items, leading to spills or broken objects. This situation could be especially concerning if the table is used as a computer desk, which might have electronic devices and equipment that could be damaged due to falling items or liquid spills. In addition, the cats' fur might cause allergies for some people or cause contamination of hygienic surfaces. To prevent accidents and maintain cleanliness, it would be best to remove the cats from the table and place them in a designated area, away from critical items or surfaces.",
"role": "assistant"
},
{
"content": "For the image on the left-hand side, what does the man need to consider while attempting the catch?",
"role": "user"
},
{
"content": "While attempting to catch the frisbee, the man should focus on various factors such as his hand-eye coordination, body position, timing, and speed. Hand-eye coordination is essential for accurately judging the trajectory of the frisbee and adjusting his movement accordingly. He should position his body in a way that maximizes his chance to grasp the frisbee efficiently. This may involve reaching up or leaping, depending on the height of the incoming frisbee. Timing is equally critical since catching the frisbee requires precise synchronization between hand movement and the arrival of the frisbee. Lastly, the man should maintain an appropriate speed to ensure he can meet the frisbee at its expected position while taking into account the frisbee's speed and spin. By considering and adapting to these factors, the man should have a better chance of successfully catching the frisbee.",
"role": "assistant"
}
] |
llava_665k_merged
|
iconqa_train_choose_img_grade2_U7_53503
|
[
"iconqa/iconqa_data/iconqa/train/choose_img/29813/image.png",
"iconqa/iconqa_data/iconqa/train/choose_img/29813/choice_0.png",
"iconqa/iconqa_data/iconqa/train/choose_img/29813/choice_1.png",
"iconqa/iconqa_data/iconqa/train/choose_img/29813/choice_2.png",
"iconqa/iconqa_data/iconqa/train/choose_img/29813/choice_3.png"
] |
[
{
"content": "<image>Which is shaped like a cube?\nA. <image>\nB. <image>\nC. <image>\nD. <image>",
"role": "user"
},
{
"content": "Answer: C",
"role": "assistant"
}
] |
iconqa
|
docvqa_xjgg0065_question_12376
|
[
"docvqa/xjgg0065_1.png"
] |
[
{
"content": "<image>how many number of pages are mentioned in the sheet?",
"role": "user"
},
{
"content": "4",
"role": "assistant"
}
] |
docvqa
|
Lin-Chen/ShareGPT4V-9306-contrastive_caption_select-35161-35167
|
[
"contrastive_caption/sharegpt4v/coco/train2017/000000174794.jpg",
"contrastive_caption/sharegpt4v/coco/train2017/000000174802.jpg",
"contrastive_caption/sharegpt4v/coco/train2017/000000174806.jpg",
"contrastive_caption/sharegpt4v/coco/train2017/000000174814.jpg",
"contrastive_caption/sharegpt4v/coco/train2017/000000174831.jpg",
"contrastive_caption/sharegpt4v/coco/train2017/000000174838.jpg"
] |
[
{
"content": "<image> <image> <image> <image> <image> <image> Which image is the caption connected to?\nIn the heart of an outdoor arena, two horses and their riders are captured in a moment of serene activity. The foreground is dominated by a chestnut-colored horse, its muscles rippling under a glossy coat as it trots with purpose. Astride this majestic creature is a rider, clad in a vibrant pink shirt and a protective black helmet, their hands expertly guiding the horse with the reins.\n\nA little further back, another scene unfolds. A palomino horse, its golden coat gleaming in the light, walks at a leisurely pace. Its rider, dressed in a calming blue shirt and a matching black helmet, holds the reins with an air of relaxed confidence.\n\nThe arena beneath them is a sandy expanse, bordered by white fences that stand out against the natural backdrop. Beyond these fences, lush trees sway gently, their green leaves rustling in the breeze. A building can be seen in the distance, its structure adding a touch of civilization to this tranquil setting.\n\nEach element in this image - from the horses and riders to the arena and its surroundings - paints a vivid picture of a peaceful day spent in harmony with nature and these magnificent creatures.",
"role": "user"
},
{
"content": "Image 6",
"role": "assistant"
},
{
"content": "Which image is the caption describing?\nIn the heart of the image, a vibrant yellow boat floats on the tranquil blue water, carrying a group of people dressed in colorful attire. Two sturdy brown cows, adorned with red blankets, are positioned on either side of the boat, pulling it along with their strength. The passengers on the boat are mostly seated, enjoying the journey, while one individual stands, perhaps guiding the cows or steering the boat. In the distance, towards the right side of the image, a small boat can be seen, adding depth to this picturesque scene. The image captures a unique moment of harmony between humans and animals, set against the backdrop of serene waters.",
"role": "user"
},
{
"content": "The fifth image",
"role": "assistant"
},
{
"content": "Which image is the caption connected to?\nThis image captures the vibrant energy of a bustling outdoor market in Africa. The market is a riot of colors, with at least eight umbrellas in hues of red, blue, green, and yellow scattered throughout the scene. These umbrellas, some in the foreground and others in the background, provide shade for the vendors and their goods.\n\nThe stalls are brimming with a variety of items, including fresh fruits and vegetables, and clothing. At least four vendors can be seen behind their stalls, engaging with customers and managing their goods. Their actions breathe life into the market scene, making it more than just a place of commerce but a hub of social interaction.\n\nThe market is set against a backdrop of a clear blue sky and a few buildings, further emphasizing its outdoor setting. The relative positions of the objects and people create a sense of depth and perspective in the image, drawing the viewer's eye from the colorful umbrellas in the foreground to the buildings in the distance.",
"role": "user"
},
{
"content": "The initial image",
"role": "assistant"
},
{
"content": "Which image is the caption connected to?\nThis image captures a vibrant street scene in San Francisco's Chinatown. The perspective is from the corner of Sacramento and Grant, looking down Sacramento towards the Bay. The street is a kaleidoscope of colors with buildings and shops painted in various hues. \n\nA white bus is making a turn onto Sacramento from Grant, adding a dynamic element to the scene. Two cars are parked on the side of the street, their metallic bodies gleaming under the sunlight. A pedestrian is seen on the sidewalk, adding a human element to the urban landscape.\n\nOn the corner, a man with a backpack stands, perhaps waiting for someone or simply taking in the sights. His casual stance contrasts with the hustle and bustle of the city around him. In the foreground, a parking meter stands like a silent sentinel, a common sight in busy city streets.\n\nEvery object and person in this image contributes to the overall narrative of city life - bustling, colorful, and full of stories waiting to be told.",
"role": "user"
},
{
"content": "The fourth image in sequence",
"role": "assistant"
},
{
"content": "Which image is the caption connected to?\nThe image captures a moment frozen in time at 12:30, as indicated by the gold hands of a black clock face affixed to the side of a dark brick building. The clock face, adorned with gold numbers, stands out against the black background, creating a striking contrast. \n\nTo the right of the clock, the building extends upwards into a tower, complete with arched windows that add an element of architectural interest. The tower is capped with a pointed roof, giving it an almost castle-like appearance. \n\nThe scene is set against a backdrop of a clear blue sky, which adds a serene and calming element to the image. The combination of these elements creates a picture that is both visually appealing and intriguing, inviting the viewer to wonder about the history and stories contained within the walls of this building.",
"role": "user"
},
{
"content": "The third image",
"role": "assistant"
},
{
"content": "Which image is the caption related to?\nIn the heart of this image, a pristine white toilet stands as the centerpiece in a bathroom that tells a story. The toilet, with its lid closed, is adorned with a basket of three rolls of toilet paper, resting peacefully on top of the toilet tank. The walls around the toilet are not just walls, they are a canvas covered in black and white newspaper clippings, each one carrying a tale of its own. \n\nAbove the toilet, demanding attention, is a biohazard sign. Its stark red and black colors contrast sharply against the monochrome backdrop of the newspaper clippings. The sign is positioned just right, so it's the first thing you see when you look at the wall.\n\nThe image captures all these elements from a slightly elevated angle, adding depth and perspective to the scene. It's as if we're looking down on this little corner of the world, privy to its unique charm and character. Every object in this image has its place, and together they create a scene that's both ordinary and extraordinary in its own way.",
"role": "user"
},
{
"content": "The final image",
"role": "assistant"
}
] |
contrastive_caption
|
train-4302-1-0
|
[
"nlvr2/train-4302-1-img0.png",
"nlvr2/train-4302-1-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nHere is a statement describing 2 images: An image contains at lest four pandas. Is it true or false? <image> <image> \nOptions:\n(A) True\n(B) False",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
dream_sim_nights_11330
|
[
"dreamsim/nights/distort/056/930_0.png",
"dreamsim/nights/distort/056/930_1.png",
"dreamsim/nights/ref/056/930.png"
] |
[
{
"content": "Answer the following question:\nIf image 3 is the reference image, which image of the other two is more similar to the reference image? <image> <image> <image>. (Assume reference image is image 1 and the other two are image 2 and image 3)\n",
"role": "user"
},
{
"content": "The image that is more similar to the reference image is image 1.",
"role": "assistant"
}
] |
dreamsim
|
train-12702-0-1
|
[
"nlvr2/train-12702-0-img0.png",
"nlvr2/train-12702-0-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\n<image> <image> Here is a statement describing these 2 images: At least one of the animals is chasing its prey. Is it true or false? \nOptions:\n(A) False\n(B) True",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
iconqa_train_choose_img_kindergarten_N7_46500
|
[
"iconqa/iconqa_data/iconqa/train/choose_img/40364/image.png",
"iconqa/iconqa_data/iconqa/train/choose_img/40364/choice_0.png",
"iconqa/iconqa_data/iconqa/train/choose_img/40364/choice_1.png",
"iconqa/iconqa_data/iconqa/train/choose_img/40364/choice_2.png"
] |
[
{
"content": "Which object is next to the baseball?<image>\nA. <image>\nB. <image>\nC. <image>",
"role": "user"
},
{
"content": "Answer: C",
"role": "assistant"
}
] |
iconqa
|
llava_665k_merged_000000200788
|
[
"llava_665k_multi/coco/train2017/000000200788.jpg"
] |
[
{
"content": "Please provide the bounding box coordinate of the region this sentence describes: car with yellow plates. <image>",
"role": "user"
},
{
"content": "[0.54, 0.52, 0.92, 0.74]",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.54, 0.52, 0.92, 0.74].",
"role": "user"
},
{
"content": "License plate.",
"role": "assistant"
},
{
"content": "Please provide the bounding box coordinate of the region this sentence describes: car with yellow plate.",
"role": "user"
},
{
"content": "[0.54, 0.52, 0.92, 0.74]",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.0, 0.34, 0.6, 0.83].",
"role": "user"
},
{
"content": "Back lights.",
"role": "assistant"
},
{
"content": "Please provide the bounding box coordinate of the region this sentence describes: the black astra with its tail lights gleaming.",
"role": "user"
},
{
"content": "[0.0, 0.34, 0.6, 0.83]",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.0, 0.34, 0.6, 0.83].",
"role": "user"
},
{
"content": "Left car.",
"role": "assistant"
}
] |
llava_665k_merged
|
197549
|
[
"vg/VG_100K_2/2389235.jpg"
] |
[
{
"content": "Identify from umbrella and woman, which object is seen the least frequently?\nPick from: A. woman, B. umbrella\n<image>",
"role": "user"
},
{
"content": "A. woman",
"role": "assistant"
}
] | null |
coinstruct_120177
|
[
"coinstruct/spaq_koniq/00327.jpg",
"coinstruct/spaq_koniq/4521235979.jpg",
"coinstruct/spaq_koniq/06913.jpg",
"coinstruct/spaq_koniq/02935.jpg"
] |
[
{
"content": "Describe and compare the quality among the four images. <image> <image> <image> <image>",
"role": "user"
},
{
"content": "Among the four images, the fourth one has the highest quality due to its clear main subject (the bird), comfortable composition, and minimal impact of blurriness on the secondary subject (the ship). The first image has average quality, with preserved texture details for the main subject but overexposure in the background. The third image's quality is passable as the main subject is identifiable despite slight blur and lack of texture detail. The second image has the poorest quality due to low sharpness, problems with focus, noise, monotonous colors, and blurry main subjects.",
"role": "assistant"
}
] |
coinstruct
|
iconqa_train_fill_in_blank_kindergarten_T7_61057
|
[
"iconqa/iconqa_data/iconqa/train/fill_in_blank/7928/image.png"
] |
[
{
"content": "Use paper clips to measure the line. The line is about (_) paper clips long.<image>",
"role": "user"
},
{
"content": "Answer: 1",
"role": "assistant"
}
] |
iconqa
|
coinstruct_114520
|
[
"coinstruct/spaq_koniq/06871.jpg",
"coinstruct/spaq_koniq/07161.jpg",
"coinstruct/spaq_koniq/09650.jpg",
"coinstruct/livefb_liveitw_aigc/000000337425_impulse_noise_3.jpg"
] |
[
{
"content": "Describe and compare the quality among the four images. <image> <image> <image> <image>",
"role": "user"
},
{
"content": "Among the four images, the fourth one has the highest quality due to its clear main subject (the bird), comfortable composition, and minimal impact of blurriness on the secondary subject (the ship). The first image has average quality, with preserved texture details for the main subject but overexposure in the background. The third image's quality is passable as the main subject is identifiable despite slight blur and lack of texture detail. The second image has the poorest quality due to low sharpness, problems with focus, noise, monotonous colors, and blurry main subjects.",
"role": "assistant"
}
] |
coinstruct
|
coinstruct_13126
|
[
"coinstruct/livefb_liveitw_aigc/midjourney_normal_052.jpg"
] |
[
{
"content": "Does the background lack detailed texture? <image>",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is the lighting sufficient in the image?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Investigate the image's attributes and provide feedback based on your critique.",
"role": "user"
},
{
"content": "The image has accurate exposure with no obvious overexposed or underexposed areas. The focus is accurate, and the image is clear with no noticeable noise. The colors are vibrant, the lighting is sufficient, and the texture details are clear. The composition is centered, and the main subject is clearly the girl in the center and the large octopus in the background, indicating good quality.",
"role": "assistant"
},
{
"content": "Are there any quality issues with this image?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Delve into the quality of the photograph and give a verdict based on your scrutiny.",
"role": "user"
},
{
"content": "The overall clarity of this image is very poor. The AI subject is relatively distinct, but lacks texture details. The overall lighting in the image is adequate, but the background lacks detailed texture. The colors in the image are rich. Therefore, the quality of this image is very poor.",
"role": "assistant"
},
{
"content": "Is the composition centered in the image?\nA. Yes\nB. No\nAnswer with the option's letter from the given choices directly.",
"role": "user"
},
{
"content": "A.",
"role": "assistant"
}
] |
coinstruct
|
train-12865-1-0
|
[
"nlvr2/train-12865-1-img0.png",
"nlvr2/train-12865-1-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nHere is a statement describing 2 images: Exactly one adult elephant with long, white ivory tusks is depicted in each image. Is it true or false? <image> <image> \nOptions:\n(A) False\n(B) True",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
chartqa_26615
|
[
"chartqa/multi_col_101085.png"
] |
[
{
"content": "What's the most popular sector for fathers?<image>",
"role": "user"
},
{
"content": "90,000 or more",
"role": "assistant"
}
] |
chartqa
|
280963
|
[
"vg/VG_100K/2328314.jpg"
] |
[
{
"content": "Which object is located on the extreme left, in the list of nose, wall, and cheese?\nOptions: (A) cheese, (B) wall, (C) nose\n<image>",
"role": "user"
},
{
"content": "(B) wall",
"role": "assistant"
}
] | null |
llava_665k_merged_000000372551-000000563869
|
[
"llava_665k_multi/coco/train2017/000000372551.jpg",
"llava_665k_multi/coco/train2017/000000563869.jpg"
] |
[
{
"content": "Considering the time visible on the clock, when is the skateboarder likely to encounter fewer crowds at the skate park, and why?\n<image>",
"role": "user"
},
{
"content": "The clock shows a time between 2 PM and 3 PM (14:00 - 15:00). Skateboarders are likely to encounter fewer crowds at the skate park during early morning or late evening hours. The reason being, most people are at work, school, or involved in other daily activities during these times of the day. Fewer crowds mean more space for the skateboarder to practice and perform tricks without distractions or interruptions. Additionally, these off-peak hours tend to help maintain a safer environment as there would be fewer people around, reducing the chances of accidents or collisions during skateboarding practice.",
"role": "assistant"
},
{
"content": "<image>\nWhy might the cat prefer to rest on the TV?",
"role": "user"
},
{
"content": "The cat might prefer to rest on the top of the flat-screen TV for several reasons. Firstly, the elevated position provides the cat with a vantage point to observe its surroundings, which can be appealing to cats as they naturally like to be up high for monitoring purposes. Secondly, the TV's surface might be warm due to heat generated by the electronic components during operation, making it an attractive spot for the cat to lay on, especially in cooler environments. Lastly, it places the cat in the center of attention, allowing it to easily watch any activity in the room and be within the line of sight of its owner or other people present, satisfying its social or attention-seeking tendencies.",
"role": "assistant"
}
] |
llava_665k_merged
|
coinstruct_49540
|
[
"coinstruct/spaq_koniq/03590.jpg",
"coinstruct/spaq_koniq/08680.jpg"
] |
[
{
"content": "Compare the quality between the first image and the second image. <image> <image>",
"role": "user"
},
{
"content": "The second image has better quality than the first image. Although the first image has good color and acceptable lighting, its composition and clarity are only average. On the other hand, the second image has a higher clarity level, which results in clearer texture details, especially in the clouds. Additionally, the scene is well-composed, which adds to the overall quality of the image. Although the colors are not very rich, and there is some loss of detail in the shadows, the positive aspects of the image outweigh the negatives, making it of higher quality compared to the first image.",
"role": "assistant"
}
] |
coinstruct
|
llava_665k_merged_2401333
|
[
"llava_665k_multi/gqa/images/2401333.jpg"
] |
[
{
"content": "<image> Is this a truck or a train?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Train",
"role": "assistant"
},
{
"content": "Do you see any cars or trains in the photograph?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Does the train car behind the other train car seem to be white?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_000000541027
|
[
"llava_665k_multi/coco/train2017/000000541027.jpg"
] |
[
{
"content": "Please provide the bounding box coordinate of the region this sentence describes: white t shirt. <image>",
"role": "user"
},
{
"content": "[0.79, 0.33, 0.94, 0.71]",
"role": "assistant"
},
{
"content": "Please provide the bounding box coordinate of the region this sentence describes: red shirt.",
"role": "user"
},
{
"content": "[0.05, 0.23, 0.35, 0.78]",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.05, 0.23, 0.35, 0.78].",
"role": "user"
},
{
"content": "Guy on left red.",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.47, 0.28, 0.71, 0.85].",
"role": "user"
},
{
"content": "Blue and red shirt.",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.05, 0.23, 0.35, 0.78].",
"role": "user"
},
{
"content": "Guy in red.",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.79, 0.33, 0.94, 0.71].",
"role": "user"
},
{
"content": "White shirt.",
"role": "assistant"
},
{
"content": "Please provide the bounding box coordinate of the region this sentence describes: guy in light jeans.",
"role": "user"
},
{
"content": "[0.47, 0.28, 0.71, 0.85]",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.05, 0.23, 0.35, 0.78].",
"role": "user"
},
{
"content": "Guy red shirt.",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.79, 0.33, 0.94, 0.71].",
"role": "user"
},
{
"content": "White shirt.",
"role": "assistant"
},
{
"content": "Please provide a short description for this region: [0.47, 0.28, 0.71, 0.85].",
"role": "user"
},
{
"content": "Guy bowling.",
"role": "assistant"
}
] |
llava_665k_merged
|
546266
|
[
"vg/VG_100K/2335258.jpg",
"vg/VG_100K_2/396.jpg",
"vg/VG_100K/2325781.jpg",
"vg/VG_100K/2332519.jpg"
] |
[
{
"content": "Tell me what attribute is consistent across these urinal across these images?\n<image> <image> <image> <image>",
"role": "user"
},
{
"content": "white",
"role": "assistant"
}
] | null |
llava_665k_merged_000000453195
|
[
"llava_665k_multi/coco/train2017/000000453195.jpg"
] |
[
{
"content": "<image> What is this kind of work setting?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Office",
"role": "assistant"
},
{
"content": "Have they been drinking beer?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "How many windows are there?",
"role": "user"
},
{
"content": "2",
"role": "assistant"
},
{
"content": "How many people are in the photo?",
"role": "user"
},
{
"content": "2",
"role": "assistant"
}
] |
llava_665k_merged
|
sharegpt4v-pt_31089
|
[
"multi_vqa/sharegpt4v/coco/train2017/000000092529.jpg",
"multi_vqa/sharegpt4v/coco/train2017/000000075637.jpg"
] |
[
{
"content": "<image><image>Which individual appears to be more prepared for a long-distance journey, the subject in image 1 or image 2?",
"role": "user"
},
{
"content": "The subject in image 1 appears to be more prepared for a long-distance journey due to the presence of the motorcycle with a bag fastened to the back, and the helmet suggesting readiness for travel.",
"role": "assistant"
},
{
"content": "If you were to create a travel story involving both subjects from image 1 and image 2, who would likely be the one traveling to meet the other, and why?",
"role": "user"
},
{
"content": "It would likely be the subject from image 1 traveling to meet the person in image 2, since the motorcycle indicates a capacity for travel, whereas the person in image 2 is engaged in a stationary recreational activity like playing with a frisbee.",
"role": "assistant"
},
{
"content": "Assuming both subjects from image 1 and image 2 are to meet at a midpoint, what mode of transport would the subject from image 1 most probably use based on the information provided?",
"role": "user"
},
{
"content": "The subject from image 1 would most probably use the motorcycle, as it is the mode of transport depicted in the image.",
"role": "assistant"
},
{
"content": "If the person in image 2 were to join the person in image 1 on a road trip, what personal item should they transfer to image 1 to secure it during the journey?",
"role": "user"
},
{
"content": "The person in image 2 should transfer the frisbee to image 1, possibly securing it in the black bag on the back of the motorcycle, to ensure it does not get lost during the road trip.",
"role": "assistant"
},
{
"content": "Which image suggests the presence of a social gathering, and what evidence supports this?",
"role": "user"
},
{
"content": "Image 2 suggests the presence of a social gathering as there are other people present in the background, albeit blurred, indicating that the person with the frisbee is not alone in the park.",
"role": "assistant"
},
{
"content": "If the person in image 2 were to navigate to the location in image 1, what kind of attire change might they consider for the journey?",
"role": "user"
},
{
"content": "The person in image 2 might consider changing into more protective gear, such as a jacket and pants suitable for riding, and a helmet similar to the one on the handlebars in image 1, to prepare for a motorcycle ride on a dirt road.",
"role": "assistant"
},
{
"content": "Combining the contexts of both images, what activity could both subjects enjoy together that would incorporate elements from each scene?",
"role": "user"
},
{
"content": "Both subjects could enjoy a motorcycle ride to a scenic park where they could take breaks for recreational activities like playing frisbee.",
"role": "assistant"
},
{
"content": "If you were to guess the time of day for both images based on the lighting and shadows, would you say they were captured at similar times or different times, and why?",
"role": "user"
},
{
"content": "Without specific information on the lighting and shadows, it is difficult to accurately guess the time of day for both images. However, if both images have similar lighting and shadow length/intensity, one might infer that they were captured at similar times of the day.",
"role": "assistant"
},
{
"content": "Considering the attire of the person in image 2, which organization might they be supporting or representing, and can you infer any specific activity related to that organization from the image?",
"role": "user"
},
{
"content": "The person in image 2 might be supporting or representing \"20.org Australia\" as indicated by the text on their blue t-shirt. While no specific activity related to the organization can be inferred, the casual and recreational setting might suggest community or outdoor events.",
"role": "assistant"
},
{
"content": "If the subject from image 1 were to take a break from their journey to watch the frisbee game in image 2, where might they safely place their helmet and why?",
"role": "user"
},
{
"content": "The subject from image 1 might safely place their helmet on the grassy area depicted in image 2, away from the frisbee play area to prevent it from getting damaged or being an obstacle during the game.",
"role": "assistant"
}
] |
synthetic_from_caption
|
llava_665k_merged_VG_100K/2356343-VG_100K/2356342
|
[
"llava_665k_multi/vg/VG_100K/2356343.jpg",
"llava_665k_multi/vg/VG_100K/2356342.jpg"
] |
[
{
"content": "<image> <image> For the image on the right-hand side, please provide a short description for this region: [0.49, 0.13, 0.81, 0.27].",
"role": "user"
},
{
"content": "Man has black hat.",
"role": "assistant"
},
{
"content": "For the image on the left, please provide a short description for this region: [0.58, 0.19, 0.68, 0.32].",
"role": "user"
},
{
"content": "White clouds in blue sky.",
"role": "assistant"
},
{
"content": "For the image two, please provide the bounding box coordinate of the region this sentence describes: black tablet near man.",
"role": "user"
},
{
"content": "[0.18, 0.55, 0.51, 0.72]",
"role": "assistant"
},
{
"content": "For the image on the right-hand side, please provide a short description for this region: [0.48, 0.14, 0.87, 0.37].",
"role": "user"
},
{
"content": "Man is wearing a hat.",
"role": "assistant"
},
{
"content": "For the first image, please provide a short description for this region: [0.58, 0.55, 0.6, 0.58].",
"role": "user"
},
{
"content": "A chimney to a building.",
"role": "assistant"
},
{
"content": "For the image one, please provide the bounding box coordinate of the region this sentence describes: A white and cloud sky.",
"role": "user"
},
{
"content": "[0.77, 0.4, 1.0, 0.57]",
"role": "assistant"
},
{
"content": "For the image on the left-hand side, please provide a short description for this region: [0.36, 0.62, 0.97, 0.83].",
"role": "user"
},
{
"content": "The building has many windows.",
"role": "assistant"
},
{
"content": "For the image on the right hand side, please provide a short description for this region: [0.0, 0.37, 0.53, 0.47].",
"role": "user"
},
{
"content": "Grey rail on board.",
"role": "assistant"
},
{
"content": "For the image on the right-hand side, please provide the bounding box coordinate of the region this sentence describes: Open laptop screen with black edge.",
"role": "user"
},
{
"content": "[0.31, 0.44, 0.54, 0.6]",
"role": "assistant"
},
{
"content": "For the image on the right-hand side, please provide the bounding box coordinate of the region this sentence describes: tablet on black case.",
"role": "user"
},
{
"content": "[0.0, 0.56, 0.49, 0.71]",
"role": "assistant"
},
{
"content": "For the image on the left hand side, please provide a short description for this region: [0.27, 0.4, 0.3, 0.43].",
"role": "user"
},
{
"content": "An airplane in the sky.",
"role": "assistant"
},
{
"content": "For the first image, please provide a short description for this region: [0.34, 0.29, 0.48, 0.36].",
"role": "user"
},
{
"content": "White airplane carrying space shuttle.",
"role": "assistant"
},
{
"content": "For the second image, please provide the bounding box coordinate of the region this sentence describes: The tablet is black.",
"role": "user"
},
{
"content": "[0.15, 0.54, 0.46, 0.66]",
"role": "assistant"
},
{
"content": "For the image two, please provide the bounding box coordinate of the region this sentence describes: the chair back is blue.",
"role": "user"
},
{
"content": "[0.0, 0.69, 0.22, 0.87]",
"role": "assistant"
},
{
"content": "For the image on the left, please provide the bounding box coordinate of the region this sentence describes: There are two jets.",
"role": "user"
},
{
"content": "[0.17, 0.4, 0.3, 0.44]",
"role": "assistant"
},
{
"content": "For the left image, please provide the bounding box coordinate of the region this sentence describes: a white airplane in the sky.",
"role": "user"
},
{
"content": "[0.17, 0.42, 0.2, 0.45]",
"role": "assistant"
},
{
"content": "For the left image, please provide the bounding box coordinate of the region this sentence describes: white clouds in blue sky.",
"role": "user"
},
{
"content": "[0.23, 0.52, 0.33, 0.6]",
"role": "assistant"
},
{
"content": "For the image on the left, please provide the bounding box coordinate of the region this sentence describes: The jets follow behind the plane.",
"role": "user"
},
{
"content": "[0.17, 0.31, 0.48, 0.44]",
"role": "assistant"
},
{
"content": "For the right image, please provide a short description for this region: [0.3, 0.44, 0.62, 0.65].",
"role": "user"
},
{
"content": "The laptop computer is open.",
"role": "assistant"
},
{
"content": "For the image two, please provide a short description for this region: [0.71, 0.3, 0.98, 0.87].",
"role": "user"
},
{
"content": "White name and number on jersey.",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_000000121601-000000025541-000000176316
|
[
"llava_665k_multi/coco/train2017/000000121601.jpg",
"llava_665k_multi/coco/train2017/000000025541.jpg",
"llava_665k_multi/coco/train2017/000000176316.jpg"
] |
[
{
"content": "For the image on the right side, what are some possible reasons for the elephants to be traveling together as a group? <image> <image> <image>",
"role": "user"
},
{
"content": "There are several possible reasons for the elephants to be traveling together as a group in this image. One reason is that elephants are social animals that typically form close-knit groups known as herds. These herds usually consist of related females and their offspring, and they offer protection, social interaction, and assistance in raising the younger members. Traveling together can also increase the group's chances of finding food and water sources, especially in a barren environment, as shown in the image. By doing so, they can pool their resources and knowledge to overcome challenges in their habitat. Additionally, traveling together provides safety from potential predators or threats, as the presence of multiple adult elephants can serve as a deterrent. It also ensures the well-being of vulnerable members like the baby elephants in the group, as they can receive support and guidance from the adults.",
"role": "assistant"
},
{
"content": "For the image one, why might one bear be observing visitors while the other one is sleeping?",
"role": "user"
},
{
"content": "The different behavior of the two bears in the image could be attributed to multiple factors, like the bears' individual personalities, their activity cycles, and/or boredom relief. Animals, including bears, may have unique temperaments that determine their level of curiosity towards their surroundings. The observant bear might be more interested in watching zoo visitors and is likely to be more active, while the sleeping bear may have a more relaxed disposition or be in a resting phase.\n\nAdditionally, as bears are known to have specific periods of activity throughout the day, it is possible that the sleeping bear is in a rest phase of its activity cycle. On the other hand, the observing bear might be in an active phase, hence showing interest in the visitors.\n\nFurthermore, zoo animals may occasionally engage in behaviors that provide them with some level of mental stimulation or ward off boredom, as their natural habitats may offer more opportunities for exploration and enrichment. Observing visitors could be one way for the bear to cope with the limited opportunities and resources within the zoo enclosure, while the sleeping bear may not be seeking any such stimulation at that moment.",
"role": "assistant"
},
{
"content": "For the image two, what is the potential reason for the elephant being in such an environment?",
"role": "user"
},
{
"content": "One potential reason for the elephant being in a barren, dusty field could be its need to find food and water resources. In the dry season, elephants may travel long distances through areas with sparse vegetation in search of food, especially if their natural habitats are affected by habitat loss or drought. Additionally, elephants also use dirt and mud to cover their bodies as a form of natural sunscreen and insect repellent. The presence of dirt in the environment shown in the image allows the elephant to engage in this behavior, protecting its skin from sunburn and repelling insects. This could be another reason for the elephant being in such an environment.",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_000000122250-000000469988-000000469988
|
[
"llava_665k_multi/coco/train2017/000000122250.jpg",
"llava_665k_multi/coco/train2017/000000469988.jpg",
"llava_665k_multi/coco/train2017/000000469988.jpg"
] |
[
{
"content": "<image> <image> <image> For the central image, please provide a short description for this region: [0.19, 0.31, 0.4, 0.65].",
"role": "user"
},
{
"content": "Large duffel backpack with red bandana tied on it.",
"role": "assistant"
},
{
"content": "For the image on the left hand side, please provide a short description for this region: [0.2, 0.68, 0.53, 0.94].",
"role": "user"
},
{
"content": "Brown stuff in bowlleft.",
"role": "assistant"
},
{
"content": "For the image on the middle side, please provide a short description for this region: [0.36, 0.49, 0.66, 0.76].",
"role": "user"
},
{
"content": "Big red thing.",
"role": "assistant"
},
{
"content": "For the image one, please provide a short description for this region: [0.2, 0.68, 0.53, 0.94].",
"role": "user"
},
{
"content": "Plate on the left.",
"role": "assistant"
},
{
"content": "For the final image, please provide the bounding box coordinate of the region this sentence describes: yellow bag.",
"role": "user"
},
{
"content": "[0.19, 0.31, 0.4, 0.65]",
"role": "assistant"
},
{
"content": "For the image on the right, please provide the bounding box coordinate of the region this sentence describes: luggage on the left.",
"role": "user"
},
{
"content": "[0.19, 0.31, 0.4, 0.65]",
"role": "assistant"
},
{
"content": "For the image on the middle side, please provide the bounding box coordinate of the region this sentence describes: blue north face backpack.",
"role": "user"
},
{
"content": "[0.4, 0.31, 0.55, 0.53]",
"role": "assistant"
},
{
"content": "For the image on the right side, please provide a short description for this region: [0.0, 0.22, 0.21, 0.45].",
"role": "user"
},
{
"content": "A turquoise fabric armed chair sitting in a living room with a folded blanket draped over the back.",
"role": "assistant"
},
{
"content": "For the image on the middle side, please provide the bounding box coordinate of the region this sentence describes: red bag.",
"role": "user"
},
{
"content": "[0.36, 0.49, 0.66, 0.76]",
"role": "assistant"
},
{
"content": "For the image on the middle, please provide a short description for this region: [0.36, 0.49, 0.66, 0.76].",
"role": "user"
},
{
"content": "A red bag infront of other bags.",
"role": "assistant"
},
{
"content": "For the first image, please provide a short description for this region: [0.54, 0.68, 0.8, 0.93].",
"role": "user"
},
{
"content": "A big bowl of broccoli pieces.",
"role": "assistant"
},
{
"content": "For the central image, please provide the bounding box coordinate of the region this sentence describes: a grey chair in a living room next tobackpacks.",
"role": "user"
},
{
"content": "[0.0, 0.31, 0.16, 0.57]",
"role": "assistant"
},
{
"content": "For the image 1, please provide a short description for this region: [0.2, 0.68, 0.53, 0.94].",
"role": "user"
},
{
"content": "Large plate with brown stuff.",
"role": "assistant"
},
{
"content": "For the image three, please provide the bounding box coordinate of the region this sentence describes: the yellow pack on left.",
"role": "user"
},
{
"content": "[0.19, 0.31, 0.4, 0.65]",
"role": "assistant"
},
{
"content": "For the image on the middle side, please provide a short description for this region: [0.19, 0.31, 0.4, 0.65].",
"role": "user"
},
{
"content": "Yellow bag.",
"role": "assistant"
},
{
"content": "For the image 3, please provide a short description for this region: [0.19, 0.31, 0.4, 0.65].",
"role": "user"
},
{
"content": "Yellow.",
"role": "assistant"
},
{
"content": "For the image 3, please provide the bounding box coordinate of the region this sentence describes: front red bag.",
"role": "user"
},
{
"content": "[0.36, 0.49, 0.66, 0.76]",
"role": "assistant"
},
{
"content": "For the image on the left, please provide a short description for this region: [0.2, 0.68, 0.53, 0.94].",
"role": "user"
},
{
"content": "The plate of meet.",
"role": "assistant"
},
{
"content": "For the image on the left, please provide a short description for this region: [0.2, 0.68, 0.53, 0.94].",
"role": "user"
},
{
"content": "The meat dish.",
"role": "assistant"
},
{
"content": "For the initial image, please provide a short description for this region: [0.54, 0.68, 0.8, 0.93].",
"role": "user"
},
{
"content": "Broccoli.",
"role": "assistant"
},
{
"content": "For the image on the right hand side, please provide the bounding box coordinate of the region this sentence describes: blue bag.",
"role": "user"
},
{
"content": "[0.4, 0.31, 0.55, 0.53]",
"role": "assistant"
},
{
"content": "For the image on the left side, please provide a short description for this region: [0.54, 0.68, 0.8, 0.93].",
"role": "user"
},
{
"content": "Front right dish.",
"role": "assistant"
},
{
"content": "For the image on the middle, please provide the bounding box coordinate of the region this sentence describes: tan bag on the left with the pink bandana tied to it.",
"role": "user"
},
{
"content": "[0.19, 0.31, 0.4, 0.65]",
"role": "assistant"
},
{
"content": "For the image on the right-hand side, please provide the bounding box coordinate of the region this sentence describes: blue tall bag.",
"role": "user"
},
{
"content": "[0.4, 0.31, 0.55, 0.53]",
"role": "assistant"
},
{
"content": "For the image one, please provide the bounding box coordinate of the region this sentence describes: plate with brown food.",
"role": "user"
},
{
"content": "[0.2, 0.68, 0.53, 0.94]",
"role": "assistant"
},
{
"content": "For the middle image, please provide the bounding box coordinate of the region this sentence describes: red bag in front.",
"role": "user"
},
{
"content": "[0.36, 0.49, 0.66, 0.76]",
"role": "assistant"
},
{
"content": "For the image on the right hand side, please provide the bounding box coordinate of the region this sentence describes: the empty grey chair on the left side of the room.",
"role": "user"
},
{
"content": "[0.0, 0.31, 0.16, 0.57]",
"role": "assistant"
},
{
"content": "For the image 1, please provide a short description for this region: [0.54, 0.68, 0.8, 0.93].",
"role": "user"
},
{
"content": "A bunch of broccoli.",
"role": "assistant"
}
] |
llava_665k_merged
|
iconqa_train_fill_in_blank_grade1_P10_63991
|
[
"iconqa/iconqa_data/iconqa/train/fill_in_blank/54894/image.png"
] |
[
{
"content": "<image>Move the ruler to measure the length of the twig to the nearest inch. The twig is about (_) inches long.",
"role": "user"
},
{
"content": "Answer: 2",
"role": "assistant"
}
] |
iconqa
|
coinstruct_65372
|
[
"coinstruct/spaq_koniq/07643.jpg",
"coinstruct/spaq_koniq/08580.jpg"
] |
[
{
"content": "Compare the quality between the first image and the second image. <image> <image>",
"role": "user"
},
{
"content": "The first image has a better quality. The reviewer specifically mentions that the texture and outline of the subject are very clear, indicating that more detail is preserved in the first image. While the second image is acceptable and has good color saturation and composition, it loses some fine texture details of the main subject and the ground, which suggests that the quality is not as high as the first image.",
"role": "assistant"
}
] |
coinstruct
|
llava_665k_merged_000000249198-000000264085-000000246345
|
[
"llava_665k_multi/coco/train2017/000000249198.jpg",
"llava_665k_multi/coco/train2017/000000264085.jpg",
"llava_665k_multi/coco/train2017/000000246345.jpg"
] |
[
{
"content": "For the first image, how many horses are pulling the carriage in the image? <image> <image> <image>",
"role": "user"
},
{
"content": "There are a couple of horses pulling the wagon in the image.",
"role": "assistant"
},
{
"content": "For the image on the right side, describe the relationship between the rider and the horse during such competitions.",
"role": "user"
},
{
"content": "In show jumping or any other equestrian sport, the relationship between the rider and the horse is of paramount importance. The rider and the horse need to have a strong bond and trust in each other's abilities to succeed in competitions. They work together as a team, with each member being responsible for their respective roles. \n\nThe rider plays an essential role in guiding, communicating, and controlling the horse throughout the course. They need to have excellent riding skills, a deep understanding of the horse's temperament, and the ability to respond promptly to any sudden changes. The rider has to use their body weight, leg pressure, and rein contact to give the horse subtle cues, or aids, which direct the horse's movements and jumps.\n\nOn the other hand, the horse must be agile, obedient, and confident in its abilities. Horses in show jumping need to possess a strong physical foundation, including balance, power, and stamina, as well as the ability to focus and adapt. They need to be well-trained and trust their rider's guidance, responding effectively to the rider's aids throughout the course. Additionally, the horse should have a strong jumping ability and a natural instinct for clearing obstacles.\n\nTogether, the rider and the horse form a partnership based on trust, communication, and mutual understanding. This essential bond allows them to navigate the course successfully, overcoming each obstacle and striving to achieve the best possible performance in the competition.",
"role": "assistant"
},
{
"content": "For the image on the right side, what is the purpose of jumping the horse over an obstacle in the image?",
"role": "user"
},
{
"content": "The purpose of jumping the horse over an obstacle is to showcase their skills in a competition, which requires the rider and the horse to successfully navigate a course with various hurdles or obstacles.",
"role": "assistant"
},
{
"content": "For the second image, what color is the dog?",
"role": "user"
},
{
"content": "The dog is black.",
"role": "assistant"
},
{
"content": "For the central image, is the dog asleep or just resting?",
"role": "user"
},
{
"content": "The dog appears to be resting or laying down, but it is not clear whether it is asleep or not from the image.",
"role": "assistant"
},
{
"content": "For the final image, is the person riding the horse a man or a woman?",
"role": "user"
},
{
"content": "The person riding the horse is a woman.",
"role": "assistant"
},
{
"content": "For the image on the right side, what kind of equestrian sport could this be?",
"role": "user"
},
{
"content": "This image represents the equestrian sport called show jumping, in which both the horse and rider are required to negotiate a course of obstacles or hurdles within a set time limit, while striving to avoid penalties for errors like knocking down rails or refusals.",
"role": "assistant"
},
{
"content": "For the image 1, what type of horses are pulling the wagon?",
"role": "user"
},
{
"content": "The Clydesdales are the type of horses pulling the wagon.",
"role": "assistant"
},
{
"content": "For the middle image, where is the dog located in the image?",
"role": "user"
},
{
"content": "The dog is located on a dog bed in a room.",
"role": "assistant"
},
{
"content": "For the image on the middle, what is the size of the dog? Small or large?",
"role": "user"
},
{
"content": "The dog is small in size.",
"role": "assistant"
},
{
"content": "For the image on the left hand side, what is the setting of the image?",
"role": "user"
},
{
"content": "The image depicts a street setting where people are watching a horse-drawn carriage go down the street during a parade.",
"role": "assistant"
},
{
"content": "For the image one, what color is the horse or horses pulling the cart?",
"role": "user"
},
{
"content": "The horse or horses pulling the cart are grey.",
"role": "assistant"
},
{
"content": "For the middle image, what color is the blanket the dog is laying on?",
"role": "user"
},
{
"content": "The dog is laying on a red blanket.",
"role": "assistant"
},
{
"content": "For the image on the left, what are the horses doing in the image?",
"role": "user"
},
{
"content": "The horses are pulling a carriage carrying a cart filled with lots of people. They are leading a parade down the street.",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_000000574747-000000574747
|
[
"llava_665k_multi/coco/train2017/000000574747.jpg",
"llava_665k_multi/coco/train2017/000000574747.jpg"
] |
[
{
"content": "<image> <image> For the image on the right side, please provide a short description for this region: [0.57, 0.33, 1.0, 0.82].",
"role": "user"
},
{
"content": "Elephant on right.",
"role": "assistant"
},
{
"content": "For the first image, please provide a short description for this region: [0.57, 0.33, 1.0, 0.82].",
"role": "user"
},
{
"content": "Eke right.",
"role": "assistant"
},
{
"content": "For the image on the left hand side, please provide a short description for this region: [0.29, 0.45, 0.49, 0.83].",
"role": "user"
},
{
"content": "Baby in front.",
"role": "assistant"
},
{
"content": "For the second image, please provide a short description for this region: [0.01, 0.25, 0.32, 0.82].",
"role": "user"
},
{
"content": "Left elephant.",
"role": "assistant"
},
{
"content": "For the image on the right-hand side, please provide a short description for this region: [0.01, 0.25, 0.32, 0.82].",
"role": "user"
},
{
"content": "Elephant looking at you.",
"role": "assistant"
},
{
"content": "For the image on the right side, please provide a short description for this region: [0.43, 0.48, 0.57, 0.77].",
"role": "user"
},
{
"content": "The right one of the two babies.",
"role": "assistant"
},
{
"content": "For the left image, please provide the bounding box coordinate of the region this sentence describes: baby elephant standing next to biggest elephant.",
"role": "user"
},
{
"content": "[0.29, 0.45, 0.49, 0.83]",
"role": "assistant"
},
{
"content": "For the right image, please provide the bounding box coordinate of the region this sentence describes: left elephant.",
"role": "user"
},
{
"content": "[0.01, 0.25, 0.32, 0.82]",
"role": "assistant"
},
{
"content": "For the image on the left hand side, please provide a short description for this region: [0.01, 0.25, 0.32, 0.82].",
"role": "user"
},
{
"content": "Far left elephant.",
"role": "assistant"
},
{
"content": "For the image on the left hand side, please provide the bounding box coordinate of the region this sentence describes: closest baby elephant.",
"role": "user"
},
{
"content": "[0.29, 0.45, 0.49, 0.83]",
"role": "assistant"
},
{
"content": "For the image on the right-hand side, please provide a short description for this region: [0.01, 0.25, 0.32, 0.82].",
"role": "user"
},
{
"content": "Biggest on the side.",
"role": "assistant"
},
{
"content": "For the image on the left hand side, please provide the bounding box coordinate of the region this sentence describes: leader.",
"role": "user"
},
{
"content": "[0.57, 0.33, 1.0, 0.82]",
"role": "assistant"
},
{
"content": "For the first image, please provide a short description for this region: [0.43, 0.48, 0.57, 0.77].",
"role": "user"
},
{
"content": "Smallest baby elephant.",
"role": "assistant"
},
{
"content": "For the image on the right hand side, please provide the bounding box coordinate of the region this sentence describes: right elephant.",
"role": "user"
},
{
"content": "[0.57, 0.33, 1.0, 0.82]",
"role": "assistant"
}
] |
llava_665k_merged
|
docvqa_znbx0223_question_15182
|
[
"docvqa/znbx0223_13.png"
] |
[
{
"content": "What is the % of Other Expenses ?<image>",
"role": "user"
},
{
"content": "22.74%",
"role": "assistant"
}
] |
docvqa
|
train-11109-2-0
|
[
"nlvr2/train-11109-2-img0.png",
"nlvr2/train-11109-2-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nHere is a statement describing 2 images: One snow plow is plowing snow. Is it true or false? <image> <image> \nOptions:\n(A) False\n(B) True",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
ImageCoDe-MSR-VTT-videoTrainValVideo_video361-shot3_0-4
|
[
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img0.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img2.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img8.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img14.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img19.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img27.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img32.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img46.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img55.jpg",
"imagecode/image-sets/MSR-VTT-videoTrainValVideo_video361-shot3_0/img62.jpg"
] |
[
{
"content": "Given this detailed description:\nThe guy on the left has most of his face and all of his body cut off from the picture. You can only see his nose and up. \nWhich image provided best matches the description?<image><image><image><image><image><image><image><image><image><image>",
"role": "user"
},
{
"content": "Answer: Image 5",
"role": "assistant"
}
] |
imagecode
|
dream_sim_nights_8102
|
[
"dreamsim/nights/distort/040/826_1.png",
"dreamsim/nights/distort/040/826_0.png",
"dreamsim/nights/ref/040/826.png"
] |
[
{
"content": "Answer the following question:\nIf image 3 is the reference image, which image of the other two is more similar to the reference image? <image> <image> <image>. (Assume reference image is image 1 and the other two are image 2 and image 3)\n",
"role": "user"
},
{
"content": "Image 1 is more similar to the reference image.",
"role": "assistant"
}
] |
dreamsim
|
llava_665k_merged_000000292730-000000554875-000000030588
|
[
"llava_665k_multi/coco/train2017/000000292730.jpg",
"llava_665k_multi/coco/train2017/000000554875.jpg",
"llava_665k_multi/coco/train2017/000000030588.jpg"
] |
[
{
"content": "<image>\nHow many candles are in this image?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "1",
"role": "assistant"
},
{
"content": "What is the woman drinking?",
"role": "user"
},
{
"content": "Water",
"role": "assistant"
},
{
"content": "Are they a couple?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "What is in her glass?",
"role": "user"
},
{
"content": "Water",
"role": "assistant"
},
{
"content": "<image>\nAre there more than 3 people in the picture?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is the rider competing?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "What is the rider holding onto?",
"role": "user"
},
{
"content": "Reins",
"role": "assistant"
},
{
"content": "<image>\nIs this a forest?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "What color are the umbrellas?",
"role": "user"
},
{
"content": "Yellow",
"role": "assistant"
},
{
"content": "Are any of the umbrellas open?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
}
] |
llava_665k_merged
|
iconqa_train_choose_img_grade1_W10_49078
|
[
"iconqa/iconqa_data/iconqa/train/choose_img/52765/image.png",
"iconqa/iconqa_data/iconqa/train/choose_img/52765/choice_0.png",
"iconqa/iconqa_data/iconqa/train/choose_img/52765/choice_1.png",
"iconqa/iconqa_data/iconqa/train/choose_img/52765/choice_2.png",
"iconqa/iconqa_data/iconqa/train/choose_img/52765/choice_3.png"
] |
[
{
"content": "<image>Which object is shaped like a cylinder?\nA. <image>\nB. <image>\nC. <image>\nD. <image>",
"role": "user"
},
{
"content": "Answer: D",
"role": "assistant"
}
] |
iconqa
|
914723
|
[
"vg/VG_100K_2/2403206.jpg",
"vg/VG_100K_2/2385367.jpg",
"vg/VG_100K/2338036.jpg"
] |
[
{
"content": "Determine the difference of attributes of school bus across these images.\nChoices: A. school bus is lined, empty, parked, clean, and yellow in Image 0, yellow in Image 1, yellow and parked in Image 2., B. school bus is yellow and parked in Image 0, lined, empty, parked, clean, and yellow in Image 1, yellow in Image 2., C. school bus is lined, empty, parked, clean, and yellow in Image 0, yellow and parked in Image 1, yellow in Image 2., D. school bus is yellow in Image 0, lined, empty, parked, clean, and yellow in Image 1, yellow and parked in Image 2.\n<image> <image> <image>",
"role": "user"
},
{
"content": "D. school bus is yellow in Image 0, lined, empty, parked, clean, and yellow in Image 1, yellow and parked in Image 2.",
"role": "assistant"
}
] | null |
train-9488-3-0
|
[
"nlvr2/train-9488-3-img0.png",
"nlvr2/train-9488-3-img1.png"
] |
[
{
"content": "Answer the following multiple-choice question:\nHere is a statement describing 2 images: One image shows a wild pig wading in brown water Is it true or false? <image> <image> \nOptions:\n(A) True\n(B) False",
"role": "user"
},
{
"content": "A",
"role": "assistant"
}
] |
nlvr2
|
coinstruct_88949
|
[
"coinstruct/spaq_koniq/00422.jpg",
"coinstruct/spaq_koniq/8975077146.jpg"
] |
[
{
"content": "<image> <image> Compare the quality between the two images.",
"role": "user"
},
{
"content": "The second image has a better quality. The clarity and fine details of the vinyl are presented very clearly, and the composition is good. There are no issues with lighting or color, and the image is not tilted. In contrast, while the first image has acceptable clarity and rich colors, the lighting is weak, the composition is not good, and the image is slightly tilted. Additionally, the people running on the track are blurry, which further reduces the image quality.",
"role": "assistant"
}
] |
coinstruct
|
iconqa_train_choose_txt_grade2_T9_7762
|
[
"iconqa/iconqa_data/iconqa/train/choose_txt/79753/image.png"
] |
[
{
"content": "Is the dotted line a line of symmetry?<image>\nA. yes\nB. no",
"role": "user"
},
{
"content": "Answer: A",
"role": "assistant"
}
] |
iconqa
|
llava_665k_merged_2368097
|
[
"llava_665k_multi/gqa/images/2368097.jpg"
] |
[
{
"content": "Who wears a ski?\nAnswer the question using a single word or phrase. <image>",
"role": "user"
},
{
"content": "Woman",
"role": "assistant"
},
{
"content": "Is the young person on the ski that is on the left?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "What is the woman on?",
"role": "user"
},
{
"content": "Ski",
"role": "assistant"
},
{
"content": "What is the person that is not old standing on?",
"role": "user"
},
{
"content": "Ski",
"role": "assistant"
},
{
"content": "What is she standing on?",
"role": "user"
},
{
"content": "Ski",
"role": "assistant"
},
{
"content": "Does the young person wear a jacket?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "Is she standing on the ski on the left?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Are there children or men in this scene?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "What is the woman standing on?",
"role": "user"
},
{
"content": "Ski",
"role": "assistant"
},
{
"content": "Who is wearing a ski?",
"role": "user"
},
{
"content": "Woman",
"role": "assistant"
},
{
"content": "What is the woman holding?",
"role": "user"
},
{
"content": "Pole",
"role": "assistant"
},
{
"content": "Who is wearing glasses?",
"role": "user"
},
{
"content": "Woman",
"role": "assistant"
},
{
"content": "This woman is holding what?",
"role": "user"
},
{
"content": "Pole",
"role": "assistant"
},
{
"content": "How big is that backpack?",
"role": "user"
},
{
"content": "Large",
"role": "assistant"
},
{
"content": "What is that woman on?",
"role": "user"
},
{
"content": "Ski",
"role": "assistant"
},
{
"content": "The woman is on what?",
"role": "user"
},
{
"content": "Ski",
"role": "assistant"
},
{
"content": "What's the woman on?",
"role": "user"
},
{
"content": "Ski",
"role": "assistant"
}
] |
llava_665k_merged
|
1137679
|
[
"vg/VG_100K/2335886.jpg"
] |
[
{
"content": "<image>\nTell me how are window and door related? Choose from: (A) pushed by, (B) posing with, (C) to the right of, (D) running through",
"role": "user"
},
{
"content": "(C) to the right of",
"role": "assistant"
}
] | null |
67719
|
[
"vg/VG_100K_2/2409457.jpg"
] |
[
{
"content": "Provide the quantity of bicycle. Selections: (A) 4, (B) 3, (C) 2, (D) 1\n<image>",
"role": "user"
},
{
"content": "(D) 1",
"role": "assistant"
}
] | null |
coinstruct_73553
|
[
"coinstruct/spaq_koniq/10844.jpg",
"coinstruct/spaq_koniq/4770991170.jpg"
] |
[
{
"content": "Which image has better quality, and why? Judge and explain briefly. <image> <image>",
"role": "user"
},
{
"content": "The second image has better quality than the first image. This is because the second image has higher clarity, with clear details of the texture on the stone wall and rocky ground surface. The composition of the second image is also well thought out, which contributes to its good image quality. On the other hand, the first image has low clarity, with lost texture details of the piled-up stones and blurry trees. Additionally, the first image has a poor composition, is slightly tilted, and has a blurry background, all of which negatively impact its quality.",
"role": "assistant"
}
] |
coinstruct
|
llava_665k_merged_738610992-826102662-B000J5REXW-62356402
|
[
"llava_665k_multi/ocr_vqa/images/738610992.jpg",
"llava_665k_multi/ocr_vqa/images/826102662.jpg",
"llava_665k_multi/ocr_vqa/images/B000J5REXW.jpg",
"llava_665k_multi/ocr_vqa/images/62356402.jpg"
] |
[
{
"content": "<image> <image> <image> <image> For the image 4, is this book related to Humor & Entertainment?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 3, who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "James Calvert",
"role": "assistant"
},
{
"content": "For the image 1, is this book related to Test Preparation?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 4, who wrote this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "C. W. Gortner",
"role": "assistant"
},
{
"content": "For the image 2, is this book related to Medical Books?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image 2, what is the title of this book?",
"role": "user"
},
{
"content": "Caregiving: A Guide for Those Who Give Care and Those Who Receive it",
"role": "assistant"
},
{
"content": "For the image 4, is this book related to Literature & Fiction?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image 4, what is the title of this book?",
"role": "user"
},
{
"content": "Mademoiselle Chanel: A Novel",
"role": "assistant"
},
{
"content": "For the image 2, who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Victoria Bumagin MSSW",
"role": "assistant"
},
{
"content": "For the image 2, what type of book is this?",
"role": "user"
},
{
"content": "Medical Books",
"role": "assistant"
},
{
"content": "For the image 2, is this book related to Self-Help?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 4, what type of book is this?",
"role": "user"
},
{
"content": "Literature & Fiction",
"role": "assistant"
},
{
"content": "For the image 3, is this book related to History?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "For the image 1, what is the title of this book?",
"role": "user"
},
{
"content": "AP® Biology Crash Course Book + Online (Advanced Placement (AP) Crash Course)",
"role": "assistant"
},
{
"content": "For the image 3, is this book related to Travel?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "For the image 3, what is the title of this book?",
"role": "user"
},
{
"content": "Fiji And The Fijians - Vol. Ii - Mission History",
"role": "assistant"
},
{
"content": "For the image 3, what type of book is this?",
"role": "user"
},
{
"content": "History",
"role": "assistant"
},
{
"content": "For the image 1, who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Michael D'Alessio",
"role": "assistant"
},
{
"content": "For the image 1, what type of book is this?",
"role": "user"
},
{
"content": "Test Preparation",
"role": "assistant"
},
{
"content": "For the image 1, is this book related to Test Preparation?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
}
] |
llava_665k_merged
|
ImageCoDe-video-storytelling-videowedding_0n3gVXmbG-I-shot13_9-1
|
[
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img482.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img487.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img493.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img500.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img508.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img514.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img526.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img540.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img556.jpg",
"imagecode/image-sets/video-storytelling-videowedding_0n3gVXmbG-I-shot13_9/img588.jpg"
] |
[
{
"content": "Given a detailed description, retrieve the target image among 10 minimally contrastive images\nDescription:\nThere is a medium size gap behind the carage and the the tree\nA. <image>\nB. <image>\nC. <image>\nD. <image>\nE. <image>\nF. <image>\nG. <image>\nH. <image>\nI. <image>\nJ. <image>",
"role": "user"
},
{
"content": "Answer: B",
"role": "assistant"
}
] |
imagecode
|
llava_665k_merged_2394392
|
[
"llava_665k_multi/gqa/images/2394392.jpg"
] |
[
{
"content": "Does the iron sculpture have small size?\nAnswer the question using a single word or phrase. <image>",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "Are there both chairs and students in this picture?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "What is the white bag made of?",
"role": "user"
},
{
"content": "Plastic",
"role": "assistant"
},
{
"content": "Is the handbag on the left side or on the right?",
"role": "user"
},
{
"content": "Left",
"role": "assistant"
},
{
"content": "Do the backpack and the floor have a different colors?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "Do you see either tables or chairs that are made of metal?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "What bag is on the floor?",
"role": "user"
},
{
"content": "Handbag",
"role": "assistant"
},
{
"content": "What is the piece of furniture to the right of the students the man is to the left of?",
"role": "user"
},
{
"content": "Chair",
"role": "assistant"
},
{
"content": "Is that floor brown or white?",
"role": "user"
},
{
"content": "White",
"role": "assistant"
},
{
"content": "Who is sitting?",
"role": "user"
},
{
"content": "Students",
"role": "assistant"
},
{
"content": "Are there any sculptures to the left of the chair?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is the backpack white or black?",
"role": "user"
},
{
"content": "White",
"role": "assistant"
},
{
"content": "Is there a black brief case or backpack?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "Are both the chair and the backpack made of the same material?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Are the backpack and the handbag the same color?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
},
{
"content": "What is the name of the item of furniture that is made of the same material as the backpack by the sculpture?",
"role": "user"
},
{
"content": "Chair",
"role": "assistant"
},
{
"content": "Who is on the sculpture made of iron?",
"role": "user"
},
{
"content": "Students",
"role": "assistant"
},
{
"content": "Who is on the sculpture in the middle?",
"role": "user"
},
{
"content": "Students",
"role": "assistant"
},
{
"content": "Do the students to the left of the other students seem to be sitting?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "On which side of the image is the white chair?",
"role": "user"
},
{
"content": "Right",
"role": "assistant"
},
{
"content": "What is the backpack made of?",
"role": "user"
},
{
"content": "Plastic",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_B00086LCP8
|
[
"llava_665k_multi/ocr_vqa/images/B00086LCP8.jpg"
] |
[
{
"content": "<image> Who is the author of this book?\nAnswer the question using a single word or phrase.",
"role": "user"
},
{
"content": "Georges Marie Haardt",
"role": "assistant"
},
{
"content": "What is the title of this book?",
"role": "user"
},
{
"content": "Across the Sahara by motor car: From Touggourt to Timbuctoo,",
"role": "assistant"
},
{
"content": "What is the genre of this book?",
"role": "user"
},
{
"content": "Travel",
"role": "assistant"
},
{
"content": "Is this book related to Travel?",
"role": "user"
},
{
"content": "Yes",
"role": "assistant"
},
{
"content": "Is this book related to Law?",
"role": "user"
},
{
"content": "No",
"role": "assistant"
}
] |
llava_665k_merged
|
coinstruct_108149
|
[
"coinstruct/spaq_koniq/3948522988.jpg",
"coinstruct/spaq_koniq/02609.jpg",
"coinstruct/spaq_koniq/08297.jpg"
] |
[
{
"content": "The first image: <image>\nThe second image: <image>\nThe third image: <image>\nDescribe and compare the quality among the three images.",
"role": "user"
},
{
"content": "When comparing the quality of the three images, the third image has the poorest quality due to serious focusing issues and backlighting that caused most of the content to be blurry. The first image has medium quality, as the sharpness is poor, the brightness is very dark, and the composition is average. The second image has lower quality than the first but higher quality than the third, as it has weak lighting, poor composition, and loss of texture details on the main subject due to reflection.",
"role": "assistant"
}
] |
coinstruct
|
416856
|
[
"vg/VG_100K_2/1952.jpg",
"vg/VG_100K/2356224.jpg"
] |
[
{
"content": "Tell me what objects are seen in all of these images?\nChoices: A. flowers, B. leaves, C. coat, D. purse\n<image> <image>",
"role": "user"
},
{
"content": "D. purse",
"role": "assistant"
}
] | null |
llava_665k_merged_000000547459-000000225566-000000225566
|
[
"llava_665k_multi/coco/train2017/000000547459.jpg",
"llava_665k_multi/coco/train2017/000000225566.jpg",
"llava_665k_multi/coco/train2017/000000225566.jpg"
] |
[
{
"content": "<image> <image> <image> For the image on the left hand side, please provide a short description for this region: [0.81, 0.26, 1.0, 0.82].",
"role": "user"
},
{
"content": "Guy far right.",
"role": "assistant"
},
{
"content": "For the image 2, please provide a short description for this region: [0.63, 0.29, 1.0, 0.86].",
"role": "user"
},
{
"content": "Kid in red.",
"role": "assistant"
},
{
"content": "For the image on the right, please provide the bounding box coordinate of the region this sentence describes: guy wearing pink.",
"role": "user"
},
{
"content": "[0.63, 0.29, 1.0, 0.86]",
"role": "assistant"
},
{
"content": "For the image on the left side, please provide a short description for this region: [0.56, 0.28, 0.86, 0.83].",
"role": "user"
},
{
"content": "Man in tan coat.",
"role": "assistant"
},
{
"content": "For the image 1, please provide the bounding box coordinate of the region this sentence describes: man on right with blue cap.",
"role": "user"
},
{
"content": "[0.81, 0.26, 1.0, 0.82]",
"role": "assistant"
},
{
"content": "For the image on the middle side, please provide the bounding box coordinate of the region this sentence describes: blue shirt.",
"role": "user"
},
{
"content": "[0.08, 0.33, 0.32, 0.67]",
"role": "assistant"
},
{
"content": "For the image on the right, please provide a short description for this region: [0.08, 0.33, 0.32, 0.67].",
"role": "user"
},
{
"content": "Blue dress.",
"role": "assistant"
},
{
"content": "For the image on the right, please provide a short description for this region: [0.08, 0.33, 0.32, 0.67].",
"role": "user"
},
{
"content": "Blue left.",
"role": "assistant"
},
{
"content": "For the image 1, please provide a short description for this region: [0.81, 0.26, 1.0, 0.82].",
"role": "user"
},
{
"content": "A black mn wearing a blue hat and blue shirt talking together.",
"role": "assistant"
},
{
"content": "For the third image, please provide a short description for this region: [0.63, 0.29, 1.0, 0.86].",
"role": "user"
},
{
"content": "Person in pink and white stripes.",
"role": "assistant"
},
{
"content": "For the image on the middle, please provide the bounding box coordinate of the region this sentence describes: the woman in the blue shirt.",
"role": "user"
},
{
"content": "[0.08, 0.33, 0.32, 0.67]",
"role": "assistant"
},
{
"content": "For the image 1, please provide the bounding box coordinate of the region this sentence describes: man in gray shirt looking at an elephant.",
"role": "user"
},
{
"content": "[0.56, 0.28, 0.86, 0.83]",
"role": "assistant"
},
{
"content": "For the image on the middle side, please provide a short description for this region: [0.08, 0.33, 0.32, 0.67].",
"role": "user"
},
{
"content": "Girl in blue.",
"role": "assistant"
},
{
"content": "For the image on the left side, please provide a short description for this region: [0.56, 0.28, 0.86, 0.83].",
"role": "user"
},
{
"content": "Front baldy.",
"role": "assistant"
},
{
"content": "For the image three, please provide a short description for this region: [0.63, 0.29, 1.0, 0.86].",
"role": "user"
},
{
"content": "Pink boy.",
"role": "assistant"
},
{
"content": "For the image on the left-hand side, please provide a short description for this region: [0.56, 0.28, 0.86, 0.83].",
"role": "user"
},
{
"content": "Tan jacket.",
"role": "assistant"
},
{
"content": "For the image on the left hand side, please provide the bounding box coordinate of the region this sentence describes: man with blue.",
"role": "user"
},
{
"content": "[0.81, 0.26, 1.0, 0.82]",
"role": "assistant"
},
{
"content": "For the central image, please provide the bounding box coordinate of the region this sentence describes: person to right.",
"role": "user"
},
{
"content": "[0.63, 0.29, 1.0, 0.86]",
"role": "assistant"
},
{
"content": "For the image on the right, please provide a short description for this region: [0.63, 0.29, 1.0, 0.86].",
"role": "user"
},
{
"content": "Red stripes.",
"role": "assistant"
},
{
"content": "For the image on the middle, please provide the bounding box coordinate of the region this sentence describes: blue shirt girl.",
"role": "user"
},
{
"content": "[0.08, 0.33, 0.32, 0.67]",
"role": "assistant"
}
] |
llava_665k_merged
|
llava_665k_merged_000000039012
|
[
"llava_665k_multi/coco/train2017/000000039012.jpg"
] |
[
{
"content": "<image> What considerations might have led city planners to place the fire hydrant in this location?",
"role": "user"
},
{
"content": "When deciding on the placement of the fire hydrant in the middle of a grassy park, city planners might have considered several factors. Proximity to nearby structures or buildings is essential, ensuring that firefighters can easily access the hydrant during emergencies. Having a fire hydrant in the park can provide an additional emergency water source for nearby areas while enhancing safety for park visitors.\n\nAnother consideration is the aesthetic appeal and preservation of the park's natural beauty. By placing the hydrant in the grass near some rocks, planners have facilitated its integration with the surrounding landscape, minimizing any visual disruption. Additionally, positioning the hydrant away from sidewalks and tree roots may prevent potential obstructions or damage to the infrastructure over time. Overall, city planners aimed to balance functionality, safety, and aesthetics when choosing this hydrant's location.",
"role": "assistant"
}
] |
llava_665k_merged
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.