2 63 1678

Chuanming Liu

Chuanming

Chuanming

AI & ML interests

Artificial Intelligence, AGI, NLP, LLMs, Multimodality, MLSys. Python/Golang/C/C++/Shell/awk&sed

Recent Activity

upvoted a collection about 17 hours ago

Foundation Text-Generation Models Below 360M Parameters

liked a dataset 1 day ago

open-r1/OpenR1-Math-220k

liked a dataset 1 day ago

omlab/VLM-R1

View all activity

Organizations

Chuanming's activity

upvoted a collection about 17 hours ago

Foundation Text-Generation Models Below 360M Parameters

Collection

Great candidates for fine-tuning targeting Wllama and Transformers.js for mobile devices, ordered by number of parameters. • 30 items • Updated 2 days ago • 22

liked 2 datasets 1 day ago

open-r1/OpenR1-Math-220k

Viewer • Updated 5 days ago • 450k • 19.6k • 401

omlab/VLM-R1

Preview • Updated 4 days ago • 2.44k • 5

liked a model 1 day ago

omlab/Qwen2.5VL-3B-VLM-R1-REC-500steps

Updated 3 days ago • 70 • 6

reacted to tianchez's post with 👍 1 day ago

Post

3844

Introducing VLM-R1!

GRPO has helped DeepSeek R1 to learn reasoning. Can it also help VLMs perform stronger for general computer vision tasks?

The answer is YES and it generalizes better than SFT. We trained Qwen 2.5 VL 3B on RefCOCO (a visual grounding task) and eval on RefCOCO Val and RefGTA (an OOD task).

https://github.com/om-ai-lab/VLM-R1