Local Preference Learning for Reasoning Compression
The story so far
LoCo, a local preference learning method for reasoning compression, teaches reasoning models when to stop reasoning, significantly reducing output length in mathematical and code tasks while maintaining or improving accuracy. The method reduces completion length by 41.7%--45.4% on mathematical tasks and 23.4%--36.3% on code tasks, achieving the best AUC on all code benchmarks. Developer Muyao Yuan states that LoCo's code and model will be publicly released.
AI-written from the reports · Updated 14 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersThink Less, Locally: Local Preference Learning for Reasoning Compression
LoCo teaches reasoning models when to stop reasoning through local preference learning, significantly reducing output length in mathematical and code tasks while maintaining or improving accuracy. The method reduces completion length by 41.7%--45.4% on mathematical tasks and 23.4%--36.3% on code tasks, while maintaining or improving accuracy, and achieves the best AUC on all code benchmarks. LoCo's code and models will be publicly released.