paper-with-me

홈 › Papers

Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration

2025-08-19 · Zhicheng Yang, Zhijiang Guo, Yinya Huang, Yongxin Wang, Dongchun Xie, Hanhui Li, Yiwei Wang, Xiaodan Liang, Jing Tang arxiv

Reinforcement Learning with Verifiable Reward (RLVR) is a powerful method for enhancing the reasoning abilities of Large Language Models, but its full potential is limited by a lack of exploration in two key areas: Depth (the difficulty of problems) and Breadth (the number of training instances). Our analysis of the popular GRPO algorithm reveals a bias that down-weights difficult, low-accuracy problems, which are crucial for improving reasoning skills. To address this, we introduce Difficulty Adaptive Rollout Sampling (DARS), a method that re-weights difficult problems by using targeted, multi-stage rollouts. DARS increases the number of rollout outcomes for these harder problems according to our proposed re-balancing schedules and leads to consistent gains in Pass@K. We discovered that increasing rollout size alone does not improve performance and may actually impair it. In contrast, scaling the batch size to increase breadth via full-batch updates significantly boosted Pass@1 metrics. This improvement stems from higher token-level entropy, ensuring robust exploration and minimized gradient noise. We further present DARS-Breadth, a combined approach that uses DARS with a large breadth of training data. This method demonstrates simultaneous gains in both Pass@K and Pass@1, confirming that depth (adaptive exploration) and breadth (scaling iteration instances) are orthogonal and complementary dimensions for unlocking the full power of RLVR.

📄 PDF Abstract BibTeX arXiv:2508.13755

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Beyond Length Scaling: Synergizing Breadth and Depth for Generative Reward Models

2026-03-02 · Qiyuan Zhang, Yufei Wang, Tianhe Wu, Can Xu 외 arxiv

Recent advancements in Generative Reward Models (GRMs) have demonstrated that scaling the length of Chain-of-Thought (CoT) reasoning considerably enhances the reliability of evaluation. However, current works predominant…

Reinforcement Learning

Beyond Pass@k: Breadth-Depth Metrics for Reasoning Boundaries

2025-10-09 · Marius Dragoi, Ioana Pintilie, Florin Gogianu, Florin Brad arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a powerful paradigm to improve Large Language Models on reasoning tasks such as coding, math or logic. To assess the reasoning boundary (the fraction o…

Reinforcement Learning

Decouple before Integration: Test-time Synthesis of SFT and RLVR Task Vectors

2026-05-01 · Chaohao Yuan, Chenghao Xiao, Yu Rong, Hong Cheng 외 arxiv

SFT and RLVR represent two fundamental yet distinct paradigms for LLM post-training, each excelling in distinct dimensions. SFT expands knowledge breadth while RLVR enhances reasoning depth. Yet integrating these complem…

Mathematical Reasoning

R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?

2025-10-09 · Yi Lu, Jianing Wang, Linsen Guo, Wei He 외 arxiv

Recent trends in test-time scaling for reasoning models (e.g., OpenAI o1, DeepSeek-R1) have led to remarkable improvements through long Chain-of-Thought (CoT). However, existing benchmarks mainly focus on immediate, sing…

Reinforcement Learning

Rethinking Multiple-Choice Questions for RLVR: Unlocking Potential via Distractor Design

2026-03-13 · Xu Guo, Qiming Ge, Jian Tong, Kedi Chen 외 arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) significantly enhances the reasoning capabilities of Large Language Models. When applied to RLVR, Multiple-Choice Questions (MCQs) offer a scalable source of verifiab…

Reinforcement Learning