English
Related papers

Related papers: Understanding Refusal in Language Models with Spar…

200 papers

Reflection, the ability of large language models (LLMs) to evaluate and revise their own reasoning, has been widely used to improve performance on complex reasoning tasks. Yet, most prior works emphasizes designing reflective prompting…

Machine Learning · Computer Science 2025-12-12 Fu-Chieh Chang , Yu-Ting Lee , Pei-Yuan Wu

Large Language Models (LLMs) offer extensive knowledge across various domains, but they may inadvertently memorize sensitive, unauthorized, or malicious data, such as personal information in the medical and financial sectors. Machine…

Computation and Language · Computer Science 2024-10-16 YuXuan Wu , Bonaventure F. P. Dossou , Dianbo Liu

Large Language Models (LLMs) remain vulnerable to optimization-based jailbreak attacks that exploit internal gradient structure. While Sparse Autoencoders (SAEs) are widely used for interpretability, their robustness implications remain…

Machine Learning · Computer Science 2026-04-22 Ahson Saiyed , Sabrina Sadiekh , Chirag Agarwal

Large Language Models' safety-aligned behaviors, such as refusing harmful queries, can be represented by linear directions in activation space. Previous research modeled safety behavior with a single direction, limiting mechanistic…

Computation and Language · Computer Science 2025-05-28 Wenbo Pan , Zhichao Liu , Qiguang Chen , Xiangyang Zhou , Haining Yu , Xiaohua Jia

Many social science questions ask how linguistic properties causally affect an audience's attitudes and behaviors. Because text properties are often interlinked (e.g., angry reviews use profane language), we must control for possible latent…

Refusal-Aware Instruction Tuning (RAIT) enables Large Language Models (LLMs) to refuse to answer unknown questions. By modifying responses of unknown questions in the training data to refusal responses such as "I don't know", RAIT enhances…

Computation and Language · Computer Science 2024-12-23 Runchuan Zhu , Zhipeng Ma , Jiang Wu , Junyuan Gao , Jiaqi Wang , Dahua Lin , Conghui He

Research on adversarial robustness in language models is currently fragmented across applications and attacks, obscuring shared vulnerabilities. In this work, we propose unifying the study of adversarial robustness in text scoring models…

Computation and Language · Computer Science 2026-02-03 Manveer Singh Tamber , Hosna Oyarhoseini , Jimmy Lin

Large Reasoning Models (LRMs) exhibit human-like cognitive reasoning strategies (e.g. backtracking, cross-verification) during reasoning process, which improves their performance on complex tasks. Currently, reasoning strategies are…

Artificial Intelligence · Computer Science 2026-01-08 Yi Fang , Wenjie Wang , Mingfeng Xue , Boyi Deng , Fengli Xu , Dayiheng Liu , Fuli Feng

Safety-aligned large language models (LLMs) are becoming increasingly widespread, especially in sensitive applications where fairness is essential and biased outputs can cause significant harm. However, evaluating the fairness of models is…

Computation and Language · Computer Science 2026-03-19 Rom Himelstein , Amit LeVi , Brit Youngmann , Yaniv Nemcovsky , Avi Mendelson

With the rapid advancement of large language models (LLMs), the safety of LLMs has become a critical concern. Despite significant efforts in safety alignment, current LLMs remain vulnerable to jailbreaking attacks. However, the root causes…

Artificial Intelligence · Computer Science 2026-03-10 Yonghong Deng , Zhen Yang , Ping Jian , Xinyue Zhang , Zhongbin Guo , Chengzhi Li

Safety alignment approaches in large language models (LLMs) often lead to the over-refusal of benign queries, significantly diminishing their utility in sensitive scenarios. To address this challenge, we introduce FalseReject, a…

Computation and Language · Computer Science 2025-07-16 Zhehao Zhang , Weijie Xu , Fanyou Wu , Chandan K. Reddy

Pre-trained language models have achieved human-level performance on many Machine Reading Comprehension (MRC) tasks, but it remains unclear whether these models truly understand language or answer questions by exploiting statistical biases…

Computation and Language · Computer Science 2021-05-26 Jieyu Lin , Jiajie Zou , Nai Ding

Sparse autoencoders (SAEs) are commonly used to interpret the internal activations of large language models (LLMs) by mapping them to human-interpretable concept representations. While existing evaluations of SAEs focus on metrics such as…

Machine Learning · Computer Science 2026-01-26 Aaron J. Li , Suraj Srinivas , Usha Bhalla , Himabindu Lakkaraju

Intervention is one of the most representative and widely used methods for understanding the internal representations of large language models (LLMs). However, existing intervention methods are confined to linear interventions grounded in…

Computation and Language · Computer Science 2026-05-15 Sangwoo Kim

Long-context LLMs can infer objectives that are not stated explicitly. This capability is useful for reasoning over documents, code, retrieved evidence, and tool traces, but it also creates a safety risk: harmful intent can be distributed…

Computation and Language · Computer Science 2026-05-15 Yu Fu , Haz Sameen Shahgir , Huanli Gong , Zhipeng Wei , N. Benjamin Erichson , Yue Dong

Negation has been shown to be a major bottleneck for masked language models, such as BERT. However, whether this finding still holds for larger-sized auto-regressive language models (``LLMs'') has not been studied comprehensively. With the…

Computation and Language · Computer Science 2023-06-16 Thinh Hung Truong , Timothy Baldwin , Karin Verspoor , Trevor Cohn

Large language models (LLMs) are known to exhibit brittle behavior under adversarial prompts and jailbreak attacks, even after extensive alignment and fine-tuning. This fragility reflects a broader challenge of modern neural language…

Computation and Language · Computer Science 2026-02-04 Patrick Cooper , Alireza Nadali , Ashutosh Trivedi , Alvaro Velasquez

Large language models (LLMs) are now ubiquitous in user-facing applications, yet they still generate undesirable toxic outputs, including profanity, vulgarity, and derogatory remarks. Although numerous detoxification methods exist, most…

Computation and Language · Computer Science 2025-10-24 Agam Goyal , Vedant Rathi , William Yeh , Yian Wang , Yuen Chen , Hari Sundaram

Omni-modal Large Language Models (OLLMs) greatly expand LLMs' multimodal capabilities but also introduce cross-modal safety risks. However, a systematic understanding of vulnerabilities in omni-modal interactions remains lacking. To bridge…

Cryptography and Security · Computer Science 2026-02-12 Kun Wang , Zherui Li , Zhenhong Zhou , Yitong Zhang , Yan Mi , Kun Yang , Yiming Zhang , Junhao Dong , Zhongxiang Sun , Qiankun Li , Yang Liu

Sparse autoencoders provide a promising unsupervised approach for extracting interpretable features from a language model by reconstructing activations from a sparse bottleneck layer. Since language models learn many concepts, autoencoders…

Machine Learning · Computer Science 2024-06-07 Leo Gao , Tom Dupré la Tour , Henk Tillman , Gabriel Goh , Rajan Troll , Alec Radford , Ilya Sutskever , Jan Leike , Jeffrey Wu