English
Related papers

Related papers: Anti-Length Shift: Dynamic Outlier Truncation for …

200 papers

Scaling test-time compute via Long Chain-of-Thought (Long-CoT) significantly enhances reasoning capabilities, yet extended generation does not guarantee correctness: after an early wrong commitment, models may keep elaborating a…

Artificial Intelligence · Computer Science 2026-01-21 Kang Chen , Fan Yu , Junjie Nian , Shihan Zhao , Zhuoka Feng , Zijun Yao , Heng Wang , Minshen Yu , Yixin Cao

Test-time compute scaling has demonstrated the ability to improve the performance of reasoning language models by generating longer chain-of-thought (CoT) sequences. However, this increase in performance comes with a significant increase in…

Artificial Intelligence · Computer Science 2025-09-24 Adarsha Balaji , Le Chen , Rajeev Thakur , Franck Cappello , Sandeep Madireddy

Knowledge distillation transfers knowledge from a large model to a small one via task and distillation losses. In this paper, we observe a trade-off between task and distillation losses, i.e., introducing distillation loss limits the…

Computer Vision and Pattern Recognition · Computer Science 2023-07-18 Borui Zhao , Quan Cui , Renjie Song , Jiajun Liang

Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks but their performance in complex logical reasoning tasks remains unsatisfactory. Although some prompting methods, such as Chain-of-Thought, can…

Computation and Language · Computer Science 2025-02-10 Tongxuan Liu , Wenjiang Xu , Weizhe Huang , Yuting Zeng , Jiaxing Wang , Xingyu Wang , Hailong Yang , Jing Li

The integration of extensive, dynamic knowledge into Large Language Models (LLMs) remains a significant challenge due to the inherent entanglement of factual data and reasoning patterns. Existing solutions, ranging from non-parametric…

Computation and Language · Computer Science 2026-02-11 Wenxuan Xie , Yujia Wang , Xin Tan , Chaochao Lu , Xia Hu , Xuhong Wang

Recently, test-time scaling Large Language Models (LLMs) have demonstrated exceptional reasoning capabilities across scientific and professional tasks by generating long chains-of-thought (CoT). As a crucial component for developing these…

Large reasoning models (LRMs) have achieved remarkable performance in complex reasoning tasks, driven by their powerful inference-time scaling capability. However, LRMs often suffer from overthinking, which results in substantial…

Computation and Language · Computer Science 2026-04-09 Yang Xiang , Yixin Ji , Ruotao Xu , Dan Qiao , Zheming Yang , Juntao Li , Min Zhang

Reasoning LLMs show improved performance with longer chains of thought. However, recent work has highlighted their tendency to overthink, continuing to revise answers even after reaching the correct solution. We quantitatively confirm this…

Machine Learning · Computer Science 2026-04-09 Xi Wang , James McInerney , Lequn Wang , Nathan Kallus

Chain-of-Thought (CoT) reasoning has emerged as a powerful tool for enhancing the problem-solving capabilities of large language models (LLMs). However, the theoretical foundations of learning from CoT data remain underdeveloped, and…

Artificial Intelligence · Computer Science 2025-07-29 Shai Shalev-Shwartz , Amnon Shashua

Reasoning large language models (RLLMs), such as OpenAI-O3 and DeepSeek-R1, have recently demonstrated remarkable capabilities by performing structured and multi-step reasoning. However, recent studies reveal that RLLMs often suffer from…

Computation and Language · Computer Science 2025-11-10 Kaiwen Yan , Xuanqing Shi , Hongcheng Guo , Wenxuan Wang , Zhuosheng Zhang , Chengwei Qin

Large reasoning models (LRMs) achieve strong performance via extended chain-of-thought (CoT) reasoning, yet suffer from excessive token consumption and high inference latency. Existing reinforcement learning (RL) approaches for CoT…

Machine Learning · Computer Science 2026-05-19 Tingcheng Bian , Yuzhe Zhang , Jing Jin , Jinchang Luo , MingQuan Cheng , Haiwei Wang , Wenyuan Jiang , Miaohui Wang

Chain-of-thought prompting (CoT) has the potential to improve the explainability of language model reasoning. But CoT can also systematically misrepresent the factors influencing models' behavior -- for example, rationalizing answers in…

Computation and Language · Computer Science 2025-06-30 James Chua , Edward Rees , Hunar Batra , Samuel R. Bowman , Julian Michael , Ethan Perez , Miles Turpin

Without relevant human priors, neural networks may learn uninterpretable features. We propose Dynamics of Attention for Focus Transition (DAFT) as a human prior for machine reasoning. DAFT is a novel method that regularizes attention-based…

Machine Learning · Statistics 2019-12-24 Wonjae Kim , Yoonho Lee

Despite rapid advancements in large language models (LLMs), the token-level autoregressive nature constrains their complex reasoning capabilities. To enhance LLM reasoning, inference-time techniques, including…

Artificial Intelligence · Computer Science 2026-01-28 Qianyue Hao , Sibo Li , Jian Yuan , Yong Li

Recent advances in natural language processing highlight two key factors for improving reasoning in large language models (LLMs): (i) allocating more test-time compute tends to help on harder problems but often introduces redundancy in the…

Computation and Language · Computer Science 2025-11-04 Riccardo Alberghi , Elizaveta Demyanenko , Luca Biggio , Luca Saglietti

Reasoning models have demonstrated remarkable progress in solving complex and logic-intensive tasks by generating extended Chain-of-Thoughts (CoTs) prior to arriving at a final answer. Yet, the emergence of this "slow-thinking" paradigm,…

Computation and Language · Computer Science 2025-09-30 Sicheng Feng , Gongfan Fang , Xinyin Ma , Xinchao Wang

Theory of Mind (ToM) assesses whether models can infer hidden mental states such as beliefs, desires, and intentions, which is essential for natural social interaction. Although recent progress in Large Reasoning Models (LRMs) has boosted…

Artificial Intelligence · Computer Science 2026-03-05 Nanxu Gong , Haotian Li , Sixun Dong , Jianxun Lian , Yanjie Fu , Xing Xie

Chain-of-Thought (CoT) prompting has emerged as a powerful technique for enhancing language model's reasoning capabilities. However, generating long and correct CoT trajectories is challenging. Recent studies have demonstrated that Looped…

Computation and Language · Computer Science 2025-02-13 Qifan Yu , Zhenyu He , Sijie Li , Xun Zhou , Jun Zhang , Jingjing Xu , Di He

The long chain-of-thought (LongCoT) capability is central to the recent breakthroughs achieved by large language models in complex reasoning tasks. However, the accompanying issue of ''underthinking'', where models exhibit shallow reasoning…

Computation and Language · Computer Science 2025-10-23 Xichen Zhang , Sitong Wu , Haoru Tan , Shaozuo Yu , Yinghao Zhu , Ziyi He , Jiaya Jia

Large reasoning models (LRMs) can do complex reasoning via long chain-of-thought (CoT) involving cognitive strategies such as backtracking and self-correction. Recent studies suggest that some models inherently possess these long reasoning…

Computation and Language · Computer Science 2025-07-18 Yunxiang Zhang , Muhammad Khalifa , Lechen Zhang , Xin Liu , Ayoung Lee , Xinliang Frederick Zhang , Farima Fatahi Bayat , Lu Wang