cainiao33/Qwen-Cup-Multimodal-RL
Qwen Cup multimodal reasoning competition (20/337 → finals 9/20 → defense 3rd 🏆): teacher-distillation SFT + small-dose GRPO RL on Qwen3.5-0.8B, reward-eval verbatim alignment, full data engineering & ablation chain
GitHub repository with 12 stars and 0 forks.
Language: Python
Topics: competition, distillation, fine-tuning, grpo, llm, multimodal, qwen, reinforcement-learning, sft, trl