Mingyang Song's picture

Mingyang Song

hitsmy

·

AI & ML interests

LVLMs

Recent Activity

upvoted an article 14 days ago

Your MoE Model Does Not Have to Select Fixed Number of Experts

upvoted a paper 30 days ago

Condition Errors Refinement in Autoregressive Image Generation with Diffusion Loss

upvoted a paper 30 days ago

P1-VL: Bridging Visual Perception and Scientific Reasoning in Physics Olympiads

View all activity

Organizations

commented 2 papers 12 months ago

FastCuRL: Curriculum Reinforcement Learning with Progressive Context Extension for Efficient Training R1-like Reasoning Models

Paper • 2503.17287 • Published Mar 21, 2025 • 11 •

From Head to Tail: Towards Balanced Representation in Large Vision-Language Models through Adaptive Data Calibration

Paper • 2503.12821 • Published Mar 17, 2025 • 10 •

commented a paper about 1 year ago

PRMBench: A Fine-grained and Challenging Benchmark for Process-Level Reward Models

Paper • 2501.03124 • Published Jan 6, 2025 • 14 •