English
Related papers

Related papers: Stable Reasoning, Unstable Responses: Mitigating L…

200 papers

Large reasoning models (LRMs) "think" by generating structured chain-of-thought (CoT) before producing a final answer, yet they still lack the ability to reason critically about safety alignment and are easily biased when a flawed premise…

Large Language Models have shown impressive generative capabilities across diverse tasks, but their safety remains a critical concern. Existing post-training alignment methods, such as SFT and RLHF, reduce harmful outputs yet leave LLMs…

Cryptography and Security · Computer Science 2025-10-21 Zhengyue Zhao , Yingzi Ma , Somesh Jha , Marco Pavone , Patrick McDaniel , Chaowei Xiao

Large Language Models (LLMs) have demonstrated impressive mathematical reasoning capabilities, yet their performance remains brittle to minor variations in problem description and prompting strategy. Furthermore, reasoning is vulnerable to…

Computation and Language · Computer Science 2025-06-23 Sam Silver , Jimin Sun , Ivan Zhang , Sara Hooker , Eddie Kim

This paper focuses on resolving stability hallucinations (e.g., repetitive or omitted speech) in LLM-based Text-to-Speech (TTS) models by improving and leveraging the attention mechanism. First, we analyzed the alignment mechanism between…

Sound · Computer Science 2026-02-16 ShiMing Wang , ZhiHao Du , Yang Xiang , TianYu Zhao , Han Zhao , Qian Chen , XianGang Li , HanJie Guo , ZhenHua Ling

Scaling inference compute enhances reasoning in large language models (LLMs), with long chains-of-thought (CoTs) enabling strategies like backtracking and error correction. Reinforcement learning (RL) has emerged as a crucial method for…

Computation and Language · Computer Science 2025-02-06 Edward Yeo , Yuxuan Tong , Morry Niu , Graham Neubig , Xiang Yue

Despite their impressive capabilities, aligned large language models (LLMs) often generate outputs that lack diversity. What drives this consistency in the generation? We investigate this phenomenon through the lens of probability…

Computation and Language · Computer Science 2026-03-04 Chenghao Yang , Sida Li , Ari Holtzman

Large language models (LLMs) have shown tremendous success in following user instructions and generating helpful responses. Nevertheless, their robustness is still far from optimal, as they may generate significantly inconsistent responses…

Computation and Language · Computer Science 2024-03-25 Yukun Zhao , Lingyong Yan , Weiwei Sun , Guoliang Xing , Shuaiqiang Wang , Chong Meng , Zhicong Cheng , Zhaochun Ren , Dawei Yin

The evolution of Large Language Model (LLM) reasoning is bottlenecked by the scarcity of high-quality process data. While self-alignment via endogenous rewards offers a solution, mining valid supervision faces three challenges: (1) Label…

Artificial Intelligence · Computer Science 2026-05-26 Yanyu Chen , Jiyue Jiang , Dianzhi Yu , Zheng Wu , Jiahong Liu , Jiaming Han , Xiao Guo , Jinhu Qi , Yu Li , Yifei Zhang , Irwin King

Training stability is typically regarded as a prerequisite for reliable optimization in large language models. In this work, we analyze how stabilizing training dynamics affects the induced generation distribution. We show that under…

Artificial Intelligence · Computer Science 2026-02-10 Xianzhe Meng , Qiangsheng Zeng , Ling Luo , Qinghan Yang , Jiarui Hao , Wenbo Wu , Qinyu Wang , Rui Yin , Lin Qi , Renzhi Lu

Understanding the extent to which Chain-of-Thought (CoT) generations align with a large language model's (LLM) internal computations is critical for deciding whether to trust an LLM's output. As a proxy for CoT faithfulness, Lanham et al.…

Computation and Language · Computer Science 2024-06-24 Oliver Bentham , Nathan Stringham , Ana Marasović

Although large language models (LLMs) have achieved great success in vast real-world applications, their vulnerabilities towards noisy inputs have significantly limited their uses, especially in high-stake environments. In these contexts,…

Computation and Language · Computer Science 2023-07-17 Zhen Zhang , Guanhua Zhang , Bairu Hou , Wenqi Fan , Qing Li , Sijia Liu , Yang Zhang , Shiyu Chang

Safety alignment incurs safety tax that perturbs a large reasoning model's (LRM) general reasoning ability. Existing datasets used for safety alignment for an LRM are usually constructed by distilling safety reasoning traces and answers…

Artificial Intelligence · Computer Science 2026-02-03 Yingsha Xie , Tiansheng Huang , Enneng Yang , Rui Min , Wenjie Lu , Xiaochun Cao , Naiqiang Tan , Li Shen

Reinforcement learning (RL) finetuning has become a key technique for enhancing large language models (LLMs) on reasoning-intensive tasks, motivating its extension to vision-language models (VLMs). While RL-tuned VLMs improve on visual…

Machine Learning · Computer Science 2026-05-22 Rosie Zhao , Anshul Shah , Xiaoyu Zhu , Xinke Deng , Zhongyu Jiang , Yang Yang , Joerg Liebelt , Arnab Mondal

Large Language Models (LLMs) have advanced various Natural Language Processing (NLP) tasks, such as text generation and translation, among others. However, these models often generate texts that can perpetuate biases. Existing approaches to…

Computation and Language · Computer Science 2025-01-07 Shaina Raza , Oluwanifemi Bamgbose , Shardul Ghuge , Fatemeh Tavakol , Deepak John Reji , Syed Raza Bashir

Large Reasoning Language Models (LRLMs or LRMs) demonstrate remarkable capabilities in complex reasoning tasks, but suffer from significant computational inefficiencies due to overthinking phenomena. Existing efficient reasoning methods…

Artificial Intelligence · Computer Science 2025-10-13 Dongqi Zheng

Large language models (LLMs) often present answers with high apparent confidence despite lacking an explicit mechanism for reasoning about certainty or truth. While existing benchmarks primarily evaluate single-turn accuracy, truthfulness…

Computation and Language · Computer Science 2026-03-05 Mohammadreza Saadat , Steve Nemzer

Large Language Models (LLMs) often exhibit strong linguistic abilities while remaining unreliable on multi-step reasoning tasks, particularly when deployed without additional training or fine-tuning. In this work, we study inference-time…

Computation and Language · Computer Science 2026-03-24 Vinay Sharma , Manish Jain

Ensuring Large Language Model (LLM) safety remains challenging due to the absence of universal standards and reliable content validators, making it difficult to obtain effective training signals. We discover that aligned models already…

Artificial Intelligence · Computer Science 2025-10-02 Guobin Shen , Dongcheng Zhao , Haibo Tong , Jindong Li , Feifei Zhao , Yi Zeng

Aligning large language models (LLMs) with human values is crucial for safe deployment. Inference-time techniques offer granular control over generation; however, they rely on model uncertainty, meaning an internal estimate of how likely…

Computation and Language · Computer Science 2026-03-04 Mohammad Atif Quamar , Mohammad Areeb , Mikhail Kuznetsov , Muslum Ozgur Ozmen , Z. Berkay Celik

Large Language Models (LLMs) display strikingly different generalization behaviors: supervised fine-tuning (SFT) often narrows capability, whereas reinforcement-learning (RL) tuning tends to preserve it. The reasons behind this divergence…

Machine Learning · Computer Science 2026-01-01 Haoyue Bai , Yiyou Sun , Wenjie Hu , Shi Qiu , Maggie Ziyu Huan , Peiyang Song , Robert Nowak , Dawn Song