English
Related papers

Related papers: Thinking-Based Non-Thinking: Solving the Reward Ha…

200 papers

Reinforcement Learning from Verifiable Rewards (RLVR) has recently shown that large language models (LLMs) can develop their own reasoning without direct supervision. However, applications in the medical domain, specifically for question…

Machine Learning · Computer Science 2025-09-22 Mirza Farhan Bin Tarek , Rahmatollah Beheshti

Large reasoning models (LRMs) have recently achieved significant progress in complex reasoning tasks, aided by reinforcement learning with verifiable rewards. However, LRMs often suffer from overthinking, expending excessive computation on…

Artificial Intelligence · Computer Science 2025-08-19 Chuhuai Yue , Chengqi Dong , Yinan Gao , Hang He , Jiajun Chai , Guojun Yin , Wei Lin

Large Reasoning Models (LRMs) have shown remarkable capabilities in solving complex problems through reinforcement learning (RL), particularly by generating long reasoning traces. However, these extended outputs often exhibit substantial…

Computation and Language · Computer Science 2025-05-22 Wei Liu , Ruochen Zhou , Yiyun Deng , Yuzhen Huang , Junteng Liu , Yuntian Deng , Yizhe Zhang , Junxian He

Reinforcement Learning with Verifiable Reward (RLVR) effectively solves complex tasks but demands extremely long context lengths during training, leading to substantial computational costs. While multi-stage training can partially mitigate…

Machine Learning · Computer Science 2025-12-25 Xin Xu , Cliveb AI , Kai Yang , Tianhao Chen , Yang Wang , Saiyong Yang , Can Yang

Large Reasoning Models (LRMs) allocate substantial inference-time compute to Chain-of-Thought (CoT) reasoning, improving performance on mathematics, scientific QA, and tool usage. However, this introduces overthinking: LRMs often reach a…

Computation and Language · Computer Science 2026-04-21 Sangjun Song , Minjae Oh , Seungkyu Lee , Sungmin Jo , Yohan Jo

Chain-of-Thought (CoT) reasoning typically utilizes the discrete language space for thinking, which is inherently inefficient, as many generated tokens only enforce linguistic rules that are not required for reasoning. To bypass this,…

Computation and Language · Computer Science 2025-12-16 Enes Özeren , Matthias Aßenmacher

Adaptive reasoning is essential for aligning the computational effort of large language models (LLMs) with the intrinsic difficulty of problems. Current chain-of-thought methods boost reasoning ability but indiscriminately generate long…

Artificial Intelligence · Computer Science 2025-12-17 Ruofan Zhang , Bin Xia , Zhen Cheng , Cairen Jian , Minglun Yang , Ngai Wong , Yuan Cheng

Reinforcement Learning (RL) has played a central role in the recent surge of LLMs' math abilities by enabling self-improvement through binary verifier signals. In contrast, Supervised Learning (SL) is rarely considered for such…

Machine Learning · Computer Science 2026-03-03 Huayu Chen , Kaiwen Zheng , Qinsheng Zhang , Ganqu Cui , Lifan Yuan , Yin Cui , Haotian Ye , Tsung-Yi Lin , Ming-Yu Liu , Jun Zhu , Haoxiang Wang

Latent reasoning represents a new development in Transformer language models that has shown potential in compressing reasoning lengths compared to chain-of-thought reasoning. By directly passing the information-rich previous final latent…

Machine Learning · Computer Science 2025-11-27 Alex Ning , Yen-Ling Kuo , Gabe Gomes

Test-time scaling (TTS) for large language models (LLMs) has thus far fallen into two largely separate paradigms: (1) reinforcement learning (RL) methods that optimize sparse outcome-based rewards, yet suffer from instability and low sample…

Machine Learning · Computer Science 2026-02-10 Can Jin , Yang Zhou , Qixin Zhang , Hongwu Peng , Di Zhang , Zihan Dong , Marco Pavone , Ligong Han , Zhang-Wei Hong , Tong Che , Dimitris N. Metaxas

Recurrent neural networks (RNNs) with deep test-time memorization modules, such as Titans and TTT, represent a promising, linearly-scaling paradigm distinct from Transformers. While these expressive models do not yet match the peak…

Machine Learning · Computer Science 2025-11-11 Zeman Li , Ali Behrouz , Yuan Deng , Peilin Zhong , Praneeth Kacham , Mahdi Karami , Meisam Razaviyayn , Vahab Mirrokni

Reasoning in large language models has long been a central research focus, and recent studies employing reinforcement learning (RL) have introduced diverse methods that yield substantial performance gains with minimal or even no external…

Hybrid reasoning language models are commonly controlled through high-level Think/No-think instructions to regulate reasoning behavior, yet we found that such mode switching is largely driven by a small set of trigger tokens rather than the…

Computation and Language · Computer Science 2026-01-13 Wang Yang , Debargha Ganguly , Xinpeng Li , Chaoda Song , Shouren Wang , Vikash Singh , Vipin Chaudhary , Xiaotian Han

The rapid evolution of Large Language Models (LLMs) has driven a growing demand for automated, high-performance system kernels to accelerate machine learning workloads. We introduce TritonRL, a domain-specialized 8B-scale LLM for Triton…

Software Engineering · Computer Science 2026-02-10 Jiin Woo , Shaowei Zhu , Allen Nie , Zhen Jia , Yida Wang , Youngsuk Park

Reinforcement learning with verifiable rewards (RLVR) typically optimizes for outcome rewards without imposing constraints on intermediate reasoning. This leaves training susceptible to reward hacking, where models exploit loopholes (e.g.,…

Machine Learning · Computer Science 2026-04-20 Songtao Wang , Quang Hieu Pham , Fangcong Yin , Xinpeng Wang , Jocelyn Qiaochu Chen , Greg Durrett , Xi Ye

Test-time scaling has enabled Large Language Models (LLMs) to tackle complex reasoning, yet the limitations of current Chain-of-Thought (CoT) evaluation obscures whether performance gains stem from genuine reasoning or mere verbosity. To…

Artificial Intelligence · Computer Science 2026-01-08 Zhizhang Fu , Yuancheng Gu , Chenkai Hu , Hanmeng Liu , Yue Zhang

Zero Reinforcement Learning (Zero-RL) has proven to be an effective approach for enhancing the reasoning capabilities of large language models (LLMs) by directly applying reinforcement learning with verifiable rewards on pretrained models,…

Artificial Intelligence · Computer Science 2025-10-30 Yuyuan Zeng , Yufei Huang , Can Xu , Qingfeng Sun , Jianfeng Yan , Guanghui Xu , Tao Yang , Fengzong Lian

Chain-of-thought reasoning in large language models can trigger an "overthinking trap": longer rollouts raise cost and latency yet often yield unreliable accuracy gains. Existing methods use global, static controls that may suppress needed…

Computation and Language · Computer Science 2026-01-22 Hanyu Li , Jiangshan Duo , Bofei Gao , Hailin Zhang , Sujian Li , Xiaotie Deng , Liang Zhao

Large Language Models (LLMs) have demonstrated impressive reasoning capabilities in complex problem-solving tasks, sparking growing interest in their application to preference reasoning in recommendation systems. Existing methods typically…

Artificial Intelligence · Computer Science 2025-10-27 Yang Zhang , Wenxin Xu , Xiaoyan Zhao , Wenjie Wang , Fuli Feng , Xiangnan He , Tat-Seng Chua

Recent advances have shown success in eliciting strong reasoning abilities in multimodal large language models (MLLMs) through rule-based reinforcement learning (RL) with outcome rewards. However, this paradigm typically lacks supervision…

Computer Vision and Pattern Recognition · Computer Science 2026-03-18 Kaixuan Fan , Kaituo Feng , Haoming Lyu , Dongzhan Zhou , Xiangyu Yue