Peng Liu's picture

2 5

Peng Liu

P3ngLiu

·

P3ngLiu

AI & ML interests

CV, Multimodal, OVD

Recent Activity

liked a dataset about 2 months ago

omlab/VLM-R1

reacted to tianchez's post with 🚀 about 2 months ago

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

reacted to tianchez's post with 👍 about 2 months ago

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

View all activity

Organizations

Articles 2

Article

1

Trials, Errors, and Breakthroughs: Our Rocky Road to OVD SOTA with Reinforcement Learning

Article

2

Improving Object Detection through Reinforcement Learning with VLM-R1

View all Articles

models

None public yet

datasets

None public yet