中文
相关论文

相关论文: HS-STaR: Hierarchical Sampling for Self-Taught Rea…

200 篇论文

Although large language models (LLMs) have achieved remarkable performance across various tasks, they remain prone to errors. A key challenge is enabling them to self-correct. While prior research has relied on external tools or large…

计算与语言 · 计算机科学 2025-03-12 Viktor Moskvoretskii , Chris Biemann , Irina Nikishina

Large Language Models (LLMs) are increasingly described as possessing strong reasoning capabilities, supported by high performance on mathematical, logical, and planning benchmarks. However, most existing evaluations rely on aggregate…

计算与语言 · 计算机科学 2026-04-16 Md. Fahad Ullah Utsho , Mohd. Ruhul Ameen , Akif Islam , Md. Golam Rashed , Dipankar Das

Large Language Models (LLMs) have succeeded remarkably in various natural language processing (NLP) tasks, yet their reasoning capabilities remain a fundamental challenge. While LLMs exhibit impressive fluency and factual recall, their…

计算与语言 · 计算机科学 2025-05-29 Avinash Patil , Aryan Jadon

Distilling large language models (LLMs) typically involves transferring the teacher model's responses through supervised fine-tuning (SFT). However, this approach neglects the potential to distill both data (output content) and reward…

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a key method for improving Large Language Models' reasoning capabilities, yet recent evidence suggests it may paradoxically shrink the reasoning boundary rather than…

人工智能 · 计算机科学 2025-10-03 Phuc Minh Nguyen , Chinh D. La , Duy M. H. Nguyen , Nitesh V. Chawla , Binh T. Nguyen , Khoa D. Doan

Large Language Models (LLMs) are emerging as promising tools for automated reinforcement learning (RL) reward design, owing to their robust capabilities in commonsense reasoning and code generation. By engaging in dialogues with RL agents,…

人工智能 · 计算机科学 2025-04-14 Zen Kit Heng , Zimeng Zhao , Tianhao Wu , Yuanfei Wang , Mingdong Wu , Yangang Wang , Hao Dong

Self-evolving trainin--where models iteratively learn from their own outputs--has emerged as a key approach for complex reasoning tasks, addressing the scarcity of high-quality chain-of-thought data. However, its effectiveness in multimodal…

计算与语言 · 计算机科学 2025-06-09 Wei Liu , Junlong Li , Xiwen Zhang , Fan Zhou , Yu Cheng , Junxian He

Integrating free-text explanations to in-context learning of large language models (LLM) is shown to elicit strong reasoning capabilities along with reasonable explanations. In this paper, we consider the problem of leveraging the…

计算与语言 · 计算机科学 2022-10-14 Shiyang Li , Jianshu Chen , Yelong Shen , Zhiyu Chen , Xinlu Zhang , Zekun Li , Hong Wang , Jing Qian , Baolin Peng , Yi Mao , Wenhu Chen , Xifeng Yan

Reinforcement learning has become the standard for improving reasoning in large language models, yet evidence increasingly suggests that RL does not teach new strategies; it redistributes probability mass over solutions the base model…

计算与语言 · 计算机科学 2026-05-12 Ömer Faruk Akgül , Rajgopal Kannan , Willie Neiswanger , Viktor Prasanna

Small Language Models (SLMs) are a cost-effective alternative to Large Language Models (LLMs), but often struggle with complex reasoning due to their limited capacity and a tendency to produce mistakes or inconsistent answers during…

计算与语言 · 计算机科学 2025-08-19 Yuanfeng Xu , Zehui Dai , Jian Liang , Jiapeng Guan , Guangrun Wang , Liang Lin , Xiaohui Lv

Large Language Models (LLMs) have achieved strong performance on static reasoning benchmarks, yet their effectiveness as interactive agents operating in adversarial, time-sensitive environments remains poorly understood. Existing…

计算机视觉与模式识别 · 计算机科学 2026-03-11 Yang Li , Xing Chen , Yutao Liu , Gege Qi , Yanxian BI , Zizhe Wang , Yunjian Zhang , Yao Zhu

Reinforcement learning has emerged as a powerful paradigm for unlocking reasoning capabilities in language models. However, relying on sparse rewards makes this process highly sample-inefficient, as models must navigate vast search spaces…

机器学习 · 计算机科学 2026-05-11 Ilia Mahrooghi , Aryo Lotfi , Emmanuel Abbe

This paper addresses two limitations of large language models (LLMs) in solving complex problems: (1) their reasoning processes exhibit Bayesian-like stochastic generation, where each token is sampled from a context-dependent probability…

人工智能 · 计算机科学 2026-04-20 Lei Lin , Jizhao Zhu , Yong Liu , Donghong Sun , Hongbo He , Yihua Du

The rapid advancements in large Language models (LLMs) have significantly enhanced their reasoning capabilities, driven by various strategies such as multi-agent collaboration. However, unlike the well-established performance improvements…

人工智能 · 计算机科学 2026-04-23 Zihan Chen , Song Wang , Zhen Tan , Xingbo Fu , Zhenyu Lei , Peng Wang , Huan Liu , Cong Shen , Jundong Li

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…

机器学习 · 计算机科学 2026-04-14 Zhicheng Yang , Zhijiang Guo , Yinya Huang , Yongxin Wang , Dongchun Xie , Hanhui Li , Yiwei Wang , Xiaodan Liang , Jing Tang

Large pretrained models are showing increasingly better performance in reasoning and planning tasks across different modalities, opening the possibility to leverage them for complex sequential decision making problems. In this paper, we…

人工智能 · 计算机科学 2024-10-10 Martin Klissarov , Devon Hjelm , Alexander Toshev , Bogdan Mazoure

Large Language Models (LLMs) exhibit strong potential in mathematical reasoning, yet their effectiveness is often limited by a shortage of high-quality queries. This limitation necessitates scaling up computational responses through…

人工智能 · 计算机科学 2025-05-20 Jingyue Gao , Runji Lin , Keming Lu , Bowen Yu , Junyang Lin , Jianyu Chen

Tabular reasoning involves interpreting natural language queries about tabular data, which presents a unique challenge of combining language understanding with structured data analysis. Existing methods employ either textual reasoning,…

数据库 · 计算机科学 2025-04-08 Nikhil Abhyankar , Vivek Gupta , Dan Roth , Chandan K. Reddy

Recent advancements in slow thinking reasoning models have shown exceptional performance in complex reasoning tasks. However, these models often exhibit overthinking (generating redundant reasoning steps for simple problems), leading to…

机器学习 · 计算机科学 2026-01-13 Yi Shen , Jian Zhang , Jieyun Huang , Shuming Shi , Wenjing Zhang , Jiangze Yan , Ning Wang , Kai Wang , Zhaoxiang Liu , Shiguo Lian

Reinforcement learning (RL) has recently become the dominant paradigm for strengthening the reasoning abilities of large language models (LLMs). Yet the rule-based reward functions commonly used on mathematical or programming benchmarks…

人工智能 · 计算机科学 2025-09-09 Haoyang He , Zihua Rong , Kun Ji , Chenyang Li , Qing Huang , Chong Xia , Lan Yang , Honggang Zhang