paper-with-me

Papers

Entropy-Aware Branching for Improved Mathematical Reasoning

2025-03-27 · Xianzhi Li, Ethan Callanan, Xiaodan Zhu, Mathieu Sibue, Antony Papadimitriou, Mahmoud Mahfouz, Zhiqiang Ma, Xiaomo Liu

While Large Language Models (LLMs) are effectively aligned through extensive pre-training and fine-tuning, they still struggle with varying levels of uncertainty during token generation. In our investigation of mathematical reasoning, we observe that errors are more likely to arise at tokens exhibiting high entropy and variance of entropy in the model's output distribution. Based on the observation, we propose a novel approach that dynamically branches the generation process on demand instead of defaulting to the single most probable token. By exploring in parallel multiple branches stemming from high probability tokens of critical decision points, the model can discover diverse reasoning paths that might otherwise be missed. We further harness external feedback from larger models to rank and select the most coherent and accurate reasoning branch. Our experimental results on mathematical word problems and calculation questions show that this branching strategy boosts the reasoning capabilities of small LLMs up to 4.6% compared to conventional argmax decoding.

📄 PDF Abstract BibTeX arXiv:2503.21961

Code (0)

등록된 구현이 없습니다.

Tasks

Mathematical Reasoning

Similar Papers 제목 키워드 기반

Adaptive Dual Reasoner: Large Reasoning Models Can Think Efficiently by Hybrid Reasoning

2025-10-11 · Yujian Zhang, Keyu Chen, Zhifeng Shen, Ruizhi Qiao 외 arxiv

Although Long Reasoning Models (LRMs) have achieved superior performance on various reasoning scenarios, they often suffer from increased computational costs and inference latency caused by overthinking. To address these…

Reinforcement LearningMathematical Reasoning

Reinforced Efficient Reasoning via Semantically Diverse Exploration

2026-01-08 · Ziqi Zhao, Zhaochun Ren, Jiahong Zou, Liu Yang 외 arxiv

Reinforcement learning with verifiable rewards (RLVR) has proven effective in enhancing the reasoning of large language models (LLMs). Monte Carlo Tree Search (MCTS)-based extensions improve upon vanilla RLVR (e.g., GRPO…

Reinforcement LearningMathematical Reasoning

Entropy-informed Decoding: Adaptive Information-Driven Branching

2026-05-10 · Benjamin Patrick Evans, Sumitra Ganesh, Leo Ardon arxiv

Large language models (LLMs) achieve remarkable generative performance, yet their output quality is dependent on the decoding strategy. While sampling-based methods (e.g., top-k, nucleus) and search-and-select based meth…

Mathematical ReasoningCode Generation

IIB-LPO: Latent Policy Optimization via Iterative Information Bottleneck

2026-01-09 · Huilin Deng, Hongchen Luo, Yue Zhu, Long Li 외 arxiv

Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) for Large Language Model (LLM) reasoning have been hindered by a persistent challenge: exploration collapse. The semantic homogeneity of random rol…

Reinforcement LearningMathematical Reasoning

ECHO: Entropy-Confidence Hybrid Optimization for Test-Time Reinforcement Learning

2026-02-02 · Chu Zhao, Enneng Yang, Yuting Liu, Jianzhe Zhao 외 arxiv

Test-time reinforcement learning generates multiple candidate answers via repeated rollouts and performs online updates using pseudo-labels constructed by majority voting. To reduce overhead and improve exploration, prio…

Reinforcement LearningVisual Reasoning