Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Free Energy-Driven Reinforcement Learning with Adaptive Advantage Shaping for Unsupervised Reasoning in LLMs

About

Unsupervised reinforcement learning (RL) has emerged as a promising paradigm for enabling self-improvement in large language models (LLMs). However, existing unsupervised RL-based methods often lack the capacity to adapt to the model's evolving reasoning capabilities during training. Therefore, these methods can misdirect policy optimization in the absence of ground-truth supervision. To address this issue, we introduce FREIA, a novel RL-based algorithm built on two key innovations: (1) Free Energy-Driven Reward (FER) adapts rewards to balance consensus and exploration based on the Free Energy Principle. (2) Adaptive Advantage Shaping (AAS) adaptively adjusts learning signals based on the statistical characteristics of sampled rewards. Empirical evaluations on nine datasets across three reasoning tasks showcase that FREIA outperforms other unsupervised RL-based baselines. Notably, in mathematical reasoning tasks, FREIA surpasses other methods by an average of 0.5 to 3.5 points in Pass@1 using the DeepSeek-R1-Distill-Qwen-1.5B model.

Yiming Huang, Zhenbo Shi, Xin-Cheng Wen, Jichuan Zeng, Cuiyun Gao, Peiyi Han, Chuanyi Liu• 2026

Related benchmarks

TaskDatasetResultRank
Mathematical ReasoningAMC 23
Pass@1 Accuracy72.5
109
Mathematical ReasoningMinerva Math
Pass@132
33
Multi-modal ReasoningGeometry3K
Pass@136.1
11
SQL GenerationBird
Pass@156.4
11
SQL GenerationSpider
Pass@174.4
6
Showing 5 of 5 rows

Other info

Follow for update