AudioRubrics

The model from Reinforcement Learning with Evolving Rubrics as Rewards for Audio Reasoning: Qwen2.5-Omni-7B post-trained with GRPO using self-evolving, audio-grounded rubric rewards and an overthinking penalty.

• Code: https://github.com/Yu-Fangxu/AudioRubrics
• Rubric dataset: https://huggingface.co/datasets/umd-zhou-lab/AVQA-Audio-Rubrics

This is the full merged checkpoint (thinker merged back into the complete Omni model) and can be served directly with vLLM:

vllm serve umd-zhou-lab/AudioRubrics --served-model-name omni --trust-remote-code \
  --max-model-len 8192 --limit-mm-per-prompt '{"audio":1}'

See the GitHub repository for training and evaluation instructions.