paper-with-me

홈 › Papers

How You Begin is How You Reason: Driving Exploration in RLVR via Prefix-Tuned Priors

2026-05-09 · Yifan Xu, Junren Chen, Yifan Chen arxiv

Reinforcement learning with verifiable rewards (RLVR) recently thrives in large language model (LLM) reasoning tasks. However, the reward sparsity and the long reasoning horizon make effective exploration challenging. In practice, this challenge manifests as the \emph{entropy collapse} phenomenon, where RLVR improves single-rollout accuracy but fails to expand coverage on successful reasoning trajectories. Passive exploration techniques like entropy regularization tend to dismiss generation quality, resulting in noisy rollouts. In response to this issue, we propose an Information-Maximizing Augmented eXploration (IMAX) framework to train a pool of soft prefixes that reshapes the base model's prior over reasoning trajectories. Rather than relying on RL to incentivize exploration on top of the base model, each prefix acts as a trainable control knob that induces a distinct rollout distribution from the same backbone model. To encourage discovery of diverse and task-relevant reasoning behaviors, we derive an Information Maximization (InfoMax) reward to complement the verifiable rewards for RL training. IMAX is in general algorithm-agnostic and can be seamlessly integrated into existing RLVR pipelines. Experiment results have shown that across three backbone scales, IMAX consistently improves reasoning performance over standard RLVR, with gains up to 11.60\% in Pass@4 and 10.57\% in Avg@4.

📄 PDF Abstract BibTeX arXiv:2605.08817

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Where Rollouts Begin: Low-Load, High-Leverage First-Token Diversification for RLVR

2026-05-27 · Soeun Kim, Albert No arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) trains reasoning models without labeled trajectories, relying on grouped rollouts to expose the policy to alternative reasoning paths and a verifier to score them. Ro…

Reinforcement Learning

Boosting LLM Exploration via Weak-Model Guidance in RLVR

2026-08-27 · Xingyu Shen, Huishuai Zhang, Peng Li, Yinchun Wang 외 arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) significantly improves LLM reasoning but often causes a drop in policy entropy, leading to narrowed reasoning coverage and degraded pass@$k$ for large $k$. While exis…

Reinforcement Learning

Well Begun, Half Done: Reinforcement Learning with Prefix Optimization for LLM Reasoning

2025-12-17 · Yiliu Sun, Zicheng Zhao, Yang Wei, Yanfang Zhang 외 arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) significantly enhances the reasoning capability of Large Language Models (LLMs). Current RLVR approaches typically conduct training across all generated tokens, but n…

Reinforcement Learning

Training Reasoning Models on Saturated Problems via Failure-Prefix Conditioning

2026-01-28 · Minwu Kim, Safal Shrestha, Anubhav Shrestha, Keith Ross arxiv

As Reinforcement Learning with Verifiable Rewards (RLVR) substantially improves the reasoning abilities of large language models (LLMs), a new bottleneck emerges: more training problems become saturated, that is, the LLM…

Reinforcement Learning

GraphPO: Graph-based Policy Optimization for Reasoning Models

2026-06-17 · Yuliang Zhan, Xinyu Tang, Jian Li, Dandan Zheng 외 arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) has become a standard paradigm for enhancing the capability of large reasoning models. RLVR typically samples responses independently and optimizes the policy using f…

Reinforcement Learning