CohereLabsCommunity (Cohere Labs Community)

hesamation

posted an update about 9 hours ago

Post

276

this paper has been blowing up
they train an open-source multimodal LLM (InternVL3) that can compete with GPT-4o and Claude 3.5 Sonnet by:
> training text and vision on a single stage
> a novel V2PE positional encoding
> SFT & mixed preference optimization
Paper: InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models (2504.10479)
> test-time scaling

1024m

authored a paper about 20 hours ago

Improving Multilingual Capabilities with Cultural and Local Knowledge in Large Language Models While Enhancing Native Performance

Paper • 2504.09753 • Published 2 days ago • 2

1024m

authored a paper 6 days ago

Kaleidoscope: In-language Exams for Massively Multilingual Vision Evaluation

Paper • 2504.07072 • Published 6 days ago • 4

hesamation

posted an update 6 days ago

Post

7106

Google published a 69-page whitepaper on Prompt Engineering and its best practices, a must-read if you are using LLMs in production:
> zero-shot, one-shot, few-shot
> system prompting
> chain-of-thought (CoT)
> ReAct

LINK: https://www.kaggle.com/whitepaper-prompt-engineering
> code prompting
> best practices

takarajordan

posted an update 7 days ago

Post

545

🎌 Two months in, https://github.com/takara-ai/go-attention has passed 429 stars on GitHub.

We built this library at takara.ai to bring attention mechanisms and transformer layers to Go — in a form that's lightweight, clean, and dependency-free.

We’re proud to say that every part of this project reflects what we set out to do.

- Pure Go — no external dependencies, built entirely on the Go standard library
- Core support for DotProductAttention and MultiHeadAttention
- Full transformer layers with LayerNorm, feed-forward networks, and residual connections
- Designed for edge, embedded, and real-time environments where simplicity and performance matter

Thank you to everyone who has supported this so far — the stars, forks, and feedback mean a lot.

4 replies

·

jjokah

posted an update 9 days ago

Post

2316

# Video Tokenization — for efficient AI video processing

Meet 𝐕𝐢𝐝𝐓𝐨𝐤, a new open-source video tokenization technique developed by Microsoft Research to address the computational challenges of processing large volumes of video data. The core problem VidTok tackles is the inefficiency caused by redundant information in raw video pixels.

VidTok converts complex video footage into compact, structured units called tokens, making it easier and more efficient for AI systems to analyze, understand, and generate video content.

Research Paper: https://arxiv.org/abs/2412.13061
VidTok Code: https://github.com/microsoft/VidTok

prithivMLmods

posted an update 9 days ago

Post

3209

Loaded some domain-specific downstream image classification content moderation models, which is essentially the practice of monitoring and filtering user-generated content on platforms, based on SigLIP-2 Base Patch16 with newly initialized trainable parameters. 🥠

+ Age-Classification-SigLIP2 : prithivMLmods/Age-Classification-SigLIP2
[ Age range classification from 0 to 65+ years ]
+ Facial-Emotion-Detection-SigLIP2 : prithivMLmods/Facial-Emotion-Detection-SigLIP2
[ Designed to classify different facial emotions ]
+ Hand-Gesture-2-Robot : prithivMLmods/Hand-Gesture-2-Robot
[ Human Hand Gesture Classification for Robot Control ]
+ Mature-Content-Detection : prithivMLmods/Mature-Content-Detection
[ Mature [adult] or neutral content categories ]
+ Vit-Mature-Content-Detection : prithivMLmods/Vit-Mature-Content-Detection
[ Mature [adult] or neutral content categories ft. ViT]
+ Human-Action-Recognition : prithivMLmods/Human-Action-Recognition
[ Human actions including clapping, sitting, running, and more ]
+ Mirage-Photo-Classifier : prithivMLmods/Mirage-Photo-Classifier
[ Whether an image is real or AI-generated (fake) ]
+ Food-101-93M : prithivMLmods/Food-101-93M
[ Classify food images into one of 101 popular dishes ]
+ Hand-Gesture-19 : prithivMLmods/Hand-Gesture-19
[ Classify hand gesture images into different categories ]
+ Trash-Net : prithivMLmods/Trash-Net
[ Classification of trash into six distinct categories ]
+ Gender-Classifier-Mini : prithivMLmods/Gender-Classifier-Mini
[ Classify images based on gender [Male / Female] ]

🎡Collections :

+ SigLIP2 Content Filters : prithivMLmods/siglip2-content-filters-models-67f001055ec2bed56ca41f6d

hesamation

posted an update 10 days ago

Post

2809

The best researchers from Yale, Stanford, Google DeepMind, and Microsoft laid out all we know about Agents in a 264-page paper [book],

Here are some of their key findings:

They build a mapping of different agent components, such as perception, memory, and world modelling, to different regions of the human brain and compare them:

- brain is much more energy-efficient
- no genuine experience in agents
- brain learns continuously, agent is static

An agent is broken down to:
- Perception: the agent's input mechanism. can be improved with multi-modality, feedback mechanisms (e.g., human corrections), etc.
- Cognition: learning, reasoning, planning, memory. LLMs are key in this part.
- Action: agent's output and tool use.

Agentic memory is represented as:
- Sensory memory or short-term holding of inputs which is not emphasized much in agents.
- Short-term memory which is the LLM context window
- Long-term memory which is the external storage such as RAG or knowledge graphs.

The memory in agents can be improved and researched in terms of:
- increasing the amount of stored information
- how to retrieve the most relevant info
- combining context-window memory with external memory
- deciding what to forget or update in memory

The agent must simulate or predict the future states of the environment for planning and decision-making.

ai world models are much simpler than the humans' with their causal reasoning (cause-and-effect) or physical intuition.

LLM world models are mostly implicit and embedded.

EMOTIONS are a deep aspect of humans, helping them with social interactions, decision-making, or learning.

Agents must understand emotions to better interact with us.

But rather than encoding the feeling of emotions, they have a surface-level modelling of emotions.

Perception is the process by which an agent receives and interprets raw data from its surroundings.

READ PAPER: Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems (2504.01990)

prithivMLmods

posted an update 10 days ago

Post

2099

ChatGPT-4o’s image generation goes wild for a week—featuring everything from Studio Ghibli-style art and image colorization to style intermixing. Here are some examples showcasing the generation of highly detailed images from freestyle design templates. Want to know more? Check out the blog 🚀

🔗Blog : https://huggingface.co/blog/prithivMLmods/chatgpt-4o-image-gen

takarajordan

posted an update 13 days ago

Post

1551

AI research over coffee ☕️
No abstracts, just bullet points.
Start your day here: https://tldr.takara.ai

1 reply

·

MarziehFadaee

authored a paper 14 days ago

Command A: An Enterprise-Ready Large Language Model

Paper • 2504.00698 • Published 15 days ago • 23

johndang-cohere

authored a paper 14 days ago

Command A: An Enterprise-Ready Large Language Model

Paper • 2504.00698 • Published 15 days ago • 23

viraat

authored a paper 14 days ago

Command A: An Enterprise-Ready Large Language Model

Paper • 2504.00698 • Published 15 days ago • 23

hesamation

posted an update 14 days ago

Post

2681

What, How, Where, and How Well? This paper reviews test-time scaling methods and all you need to know about them:
> parallel, sequential, hybrid, internal scaling
> how to scale (SFT, RL, search, verification)
> metrics and evals of test-time scaling

🔗paper: What, How, Where, and How Well? A Survey on Test-Time Scaling in Large Language Models (2503.24235)

If you want to learn what inference-time compute scaling is @rasbt has a great blog post on that:
https://magazine.sebastianraschka.com/p/state-of-llm-reasoning-and-inference-scaling

hesamation

posted an update 15 days ago

Post

1787

this paper lists ways to make reasoning LLMs more efficient:
> enforce token limits per reasoning step
> route tasks to different models (small/large)
> compress reasoning chains during SFT
> reward based on reasoning length
> parallel search at test-time
and more...
@Xiaoye08 @yaful @Warrieryes
A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, and Beyond (2503.21614)

mmhamdy

posted an update 17 days ago

Post

1566

What inspired the Transformer architecture in the "Attention Is All You Need" paper? And how were various ideas combined to create this groundbreaking model?

In this lengthy article, I explore the story and the origins of some of the ideas introduced in the paper. We'll explore everything from the fundamental attention mechanism that lies at its heart to the surprisingly simple explanation for its name, Transformer.

💡 Examples of ideas explored in the article:

✅ What was the inspiration for the attention mechanism?
✅ How did we go from attention to self-attention?
✅ Did the team have any other names in mind for the model?

and more...

I aim to tell the story of Transformers as I would have wanted to read it, and hopefully, one that appeals to others interested in the details of this fascinating idea. This narrative draws from video interviews, lectures, articles, tweets/Xs, and some digging into the literature. I have done my best to be accurate, but errors are possible. If you find inaccuracies or have any additions, please do reach out, and I will gladly make the necessary updates.

Read the article: https://huggingface.co/blog/mmhamdy/pandemonium-the-transformers-story

prithivMLmods

posted an update 17 days ago

Post

1857

Luna, the single-speaker text-to-speech model, features a Radio & Atcosim-style sound with a female voice. It offers authentic radio podcast noise and empathetic speech generation, fine-tuned based on Orpheus's Llama-based speech generation state-of-the-art model. 🎙️

+ Model : prithivMLmods/Llama-3B-Mono-Luna
+ Collection : prithivMLmods/clean-radio-mono-voice-67e76fe1b3a87cc3bccef803
+ Reference ft : https://github.com/canopyai/Orpheus-TTS
+ Base Model : canopylabs/orpheus-3b-0.1-ft

I also tried some other clean-voice single-speaker models based on Orpheus. If you're interested, check out the collection.

🔉Try the Mono Luna demo here: http://colab.research.google.com/drive/1K0AAIOKDE5XE0znxXaiiUJvPSpFveteK

6 replies

·

Aurelien-Morgan

posted an update 18 days ago

Post

1975

Almost there !
https://test.pypi.org/project/test-010-retrain-pipelines/

prithivMLmods

posted an update 20 days ago

Post

1696

Dropping some new Journey Art and Realism adapters for Flux.1-Dev, including Thematic Arts, 2021 Memory Adapters, Thread of Art, Black of Art, and more. For more details, visit the model card on Stranger Zone HF 🤗

+ Black-of-Art-Flux : strangerzonehf/Black-of-Art-Flux
+ Thread-of-Art-Flux : strangerzonehf/Thread-of-Art-Flux
+ 2021-Art-Flux : strangerzonehf/2021-Art-Flux
+ 3d-Station-Toon : strangerzonehf/3d-Station-Toon
+ New-Journey-Art-Flux : strangerzonehf/New-Journey-Art-Flux
+ Casual-Pencil-Pro : strangerzonehf/Casual-Pencil-Pro
+ Realism-H6-Flux : strangerzonehf/Realism-H6-Flux

- Repository Page :

strangerzonehf

The best dimensions and inference settings for optimal results are as follows: A resolution of 1280 x 832 with a 3:2 aspect ratio is recommended for the best quality, while 1024 x 1024 with a 1:1 aspect ratio serves as the default option. For inference, the recommended number of steps ranges between 30 and 35 to achieve optimal output.

1 reply

·

takarajordan

posted an update 22 days ago

Post

1862

Takara takes 3rd place in the {tech:munich} AI hackathon with Fudeno!

A little over 2 weeks ago @aldigobbler and I set out to create the largest MultiModal SVG dataset ever created, we succeeded in this and when I was in Munich, Germany I took it one step further and made an entire app with it!

We fine-tuned Mistral Small, made a Next.JS application and blew some minds, taking 3rd place out of over 100 hackers. So cool!

If you want to see the dataset, please see below.

takara-ai/fudeno-instruct-4M

Cohere Labs Community

AI & ML interests

Recent Activity

CohereLabsCommunity's activity

Improving Multilingual Capabilities with Cultural and Local Knowledge in Large Language Models While Enhancing Native Performance

Kaleidoscope: In-language Exams for Massively Multilingual Vision Evaluation

Command A: An Enterprise-Ready Large Language Model

Command A: An Enterprise-Ready Large Language Model

Command A: An Enterprise-Ready Large Language Model

AI & ML interests

Recent Activity

Team members 166

CohereLabsCommunity's activity