TL;DR

A method for preserving reasoning diversity during RLVR by conditioning the target model on partial trajectories produced by a smaller, weaker model.

Why it matters

RLVR often improves pass@1 while entropy and large-k coverage decline. The paper tests whether cross-model prefixes can create useful exploration pressure without an additional supervised-training stage or a more elaborate reward.

Key findings

  1. 01

    The method uses unfamiliar weak-model prefixes to interrupt overconfident generation and expose the policy to different reasoning paths.

  2. 02

    Across the reported mathematical benchmarks, gains increase as k grows, which the authors interpret as expanded reasoning coverage.

  3. 03

    The approach is designed to mitigate entropy collapse without additional SFT, reward redesign, or complex prompting.

Scaling dimensions

Models, methods & benchmarks

Algorithms
RLVR, Weak-model guidance

Topics

Read the original sourcearXiv