Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Deep Dense Exploration for LLM Reinforcement Learning via Pivot-Driven Resampling

About

Effective exploration is a key challenge in reinforcement learning for large language models: discovering high-quality trajectories within a limited sampling budget from the vast natural language sequence space. Existing methods face notable limitations: GRPO samples exclusively from the root, saturating high-probability trajectories while leaving deep, error-prone states under-explored. Tree-based methods blindly disperse budgets across trivial or unrecoverable states, causing sampling dilution that fails to uncover rare correct suffixes and destabilizes local baselines. To address this, we propose Deep Dense Exploration (DDE), a strategy that focuses exploration on $\textit{pivots}$-deep, recoverable states within unsuccessful trajectories. We instantiate DDE with DEEP-GRPO, which introduces three key innovations: (1) a lightweight data-driven utility function that automatically balances recoverability and depth bias to identify pivot states; (2) local dense resampling at each pivot to increase the probability of discovering correct subsequent trajectories; and (3) a dual-stream optimization objective that decouples global policy learning from local corrective updates. Experiments on mathematical reasoning benchmarks demonstrate that our method consistently outperforms GRPO, tree-based methods, and other strong baselines.

Yiran Guo, Zhongjian Qiao, Yingqi Xie, Jie Liu, Dan Ye, Ruiqing Zhang, Shuang Qiu, Lijie Xu• 2026

Related benchmarks

TaskDatasetResultRank
Mathematical ReasoningGSM8K (test)
Accuracy67.7
797
Mathematical ReasoningAMC
Accuracy65.1
151
Mathematical ReasoningMinerva--
138
Mathematical ReasoningAIME 24
Accuracy46.7
113
Mathematical ReasoningMATH 500
MATH 500 Accuracy81.6
106
Mathematical ReasoningOlympiad
Accuracy (%)42.6
21
Showing 6 of 6 rows

Other info

Follow for update