Delta Activations: A Representation for Finetuned Large Language Models
Paper
•
2509.04442
•
Published
•
3
Prithiv Sakthi
prithivMLmods
AI & ML interests
computer vision, nlp, multimodality
@strangerzonehf @strangerguardhf
Recent Activity
replied to
their
post
about 3 hours ago
Dropped the HeadshotX : a super-realistic headshot adapter for https://huggingface.co/Qwen/Qwen-Image, an image generation model by Qwen. It is an advanced LoRA adaptation of the Qwen-Image model and an upgraded version of https://huggingface.co/prithivMLmods/Qwen-Image-Studio-Realism, offering more precise portrait rendering with a strong focus on realism. The model was trained on diverse face types from across the world, labeled with `florence2-en` and caption-optimized using https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B. 11(types) × 5 different face types: Asian, Hispanic, Caucasian, Latina, Middle Eastern, etc.
⮞ Model🤗: https://huggingface.co/prithivMLmods/Qwen-Image-HeadshotX
⮞ The Previous Adapter (LoRA): https://huggingface.co/prithivMLmods/Qwen-Image-Studio-Realism
⮞ Collection: https://huggingface.co/collections/prithivMLmods/qwen-image-exp-lora-68a978fe11400bc3165b0c4d
.
.
.
To know more about it, visit the app page or the respective model page!!
replied to
their
post
about 7 hours ago
OpenGVLab's InternVL3_5-2B-MPO [Mixed Preference Optimization (MPO)] is a compact vision-language model in the InternVL3.5 series. You can now experience it in the Tiny VLMs Lab, an app featuring 15+ multimodal VLMs ranging from 250M to 4B parameters. These models support tasks such as OCR, reasoning, single-shot answering with small models, and captioning (including ablated variants), across a broad range of visual categories. They are also capable of handling images with complex, sensitive, or nuanced content, while adapting to varying aspect ratios and resolutions.
✨ Space/App : https://huggingface.co/spaces/prithivMLmods/Tiny-VLMs-Lab
🫙 Model : https://huggingface.co/OpenGVLab/InternVL3_5-2B-MPO
↗️ Collection: https://huggingface.co/collections/OpenGVLab/internvl35-68ac87bd52ebe953485927fb
🗞️ Paper : https://arxiv.org/pdf/2508.18265
↗️ Multimodal Space Collection : https://huggingface.co/collections/prithivMLmods/multimodal-implementations-67c9982ea04b39f0608badb0
To learn more, visit the relevant spaces, collections, and model cards.
updated
a collection
about 7 hours ago
DeepCaption attr.
Organizations
upvoted
3
papers
about 7 hours ago
upvoted
a
collection
about 14 hours ago
upvoted
5
papers
1 day ago
Video-MTR: Reinforced Multi-Turn Reasoning for Long Video Understanding
Paper
•
2508.20478
•
Published
•
16
Transition Models: Rethinking the Generative Learning Objective
Paper
•
2509.04394
•
Published
•
19
Inverse IFEval: Can LLMs Unlearn Stubborn Training Conventions to Follow Real Instructions?
Paper
•
2509.04292
•
Published
•
45
Towards a Unified View of Large Language Model Post-Training
Paper
•
2509.04419
•
Published
•
53
Drivel-ology: Challenging LLMs with Interpreting Nonsense with Depth
Paper
•
2509.03867
•
Published
•
165
upvoted
2
papers
4 days ago
upvoted
3
papers
5 days ago
How Can Input Reformulation Improve Tool Usage Accuracy in a Complex Dynamic Environment? A Study on τ-bench
Paper
•
2508.20931
•
Published
•
15
PVPO: Pre-Estimated Value-Based Policy Optimization for Agentic Reasoning
Paper
•
2508.21104
•
Published
•
28
T2R-bench: A Benchmark for Generating Article-Level Reports from Real World Industrial Tables
Paper
•
2508.19813
•
Published
•
20
upvoted
a
collection
6 days ago
upvoted
a
collection
10 days ago
upvoted
3
papers
11 days ago
Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks
Paper
•
2508.18672
•
Published
•
9
ReportBench: Evaluating Deep Research Agents via Academic Survey Tasks
Paper
•
2508.15804
•
Published
•
14
TreePO: Bridging the Gap of Policy Optimization and Efficacy and Inference Efficiency with Heuristic Tree-based Modeling
Paper
•
2508.17445
•
Published
•
78
upvoted
a
collection
11 days ago