LoCo proposed: Local preference learning for reasoning compression
To reduce inference costs caused by long chain-of-thought in reasoning models, a method called LoCo has been proposed to determine if further reasoning is necessary.
Large reasoning models achieve strong performance on challenging tasks, but their long chain-of-thought often incurs substantial inference costs by continuing to deliberate after sufficient information has been obtained. Existing reasoning compression methods primarily operate on reasoning trajectories, paying less attention to the local decision of whether further reasoning is necessary given the current state.
In response, LoCo, a local preference learning method for reasoning compression, has been proposed. This paper received 64 upvotes on Hugging Face Daily Papers.
This item is a summary. The full, checkable version is at the original source:
Read the original at Hugging Face Papers ↗