Huang Lihong's picture

4

Huang Lihong

lihong-huang

·

AI & ML interests

None yet

Recent Activity

reacted to Jaward's post with 👍 about 1 month ago

made a few improvements on custom grpo trainer: - added sequence similarity reward (seems to work) - improved vllm support (5x inference speed) - adjusted reward scores (this helped with format/accuracy) - can now push to hf hub (already pushed mine lol: https://huggingface.co/Jaward/smollm2_360m_grpo_gsm8k_reasoner) Code: https://github.com/Jaykef/ai-algorithms/blob/main/smollm2_360M_135M_grpo_gsm8k.ipynb

published a model 2 months ago

lihong-huang/Qwen2.5-1.5B-Open-R1-GRPO

liked a Space 3 months ago

sitammeur/PicQ

View all activity

Organizations

None yet

lihong-huang's activity

liked a Space 3 months ago

PicQ

Demo for MiniCPM-o 2.6 to answer questions about images

liked a model 3 months ago

openbmb/MiniCPM-o-2_6

Any-to-Any • Updated 10 days ago • 849k • 1.08k

liked a model 5 months ago

Rodeszones/CogVLM-grounding-generalist-hf-quant4

Object Detection • Updated Aug 16, 2024 • 128 • 9

liked a model 6 months ago

mistralai/Pixtral-12B-2409

Image-Text-to-Text • Updated Dec 26, 2024 • • 627