Refusal in Language Models Is Mediated by a Single Direction

17 June 2024

Nina Panickssery

Papers citing "Refusal in Language Models Is Mediated by a Single Direction"

32 / 32 papers shown

Title
On the Limitations of Steering in Language Model Alignment Chebrolu Niranjan Kokil Jaidka G. Yeo LLMSV 29 0 0 02 May 2025
Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control Hannah Cyberey David E. Evans LLMSV 67 0 0 23 Apr 2025
Misaligned Roles, Misplaced Images: Structural Input Perturbations Expose Multimodal Alignment Blind Spots Erfan Shayegani G M Shahariar Sara Abdali Lei Yu Nael B. Abu-Ghazaleh Yue Dong AAML 34 0 0 01 Apr 2025
Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinations Ziwei Ji L. Yu Yeskendir Koishekenov Yejin Bang Anthony Hartshorn Alan Schelten Cheng Zhang Pascale Fung Nicola Cancedda 38 1 0 18 Mar 2025
Can A Society of Generative Agents Simulate Human Behavior and Inform Public Health Policy? A Case Study on Vaccine Hesitancy Abe Bohan Hou H. Du Yichen Wang Jingyu Zhang Zixiao Wang Paul Pu Liang Daniel Khashabi Lauren M. Gardner Tianxing He 36 0 0 12 Mar 2025
Using Mechanistic Interpretability to Craft Adversarial Attacks against Large Language Models Thomas Winninger Boussad Addad Katarzyna Kapusta AAML 59 0 0 08 Mar 2025
Linear Representations of Political Perspective Emerge in Large Language Models Junsol Kim James Evans Aaron Schein 56 1 0 03 Mar 2025
A generative approach to LLM harmfulness detection with special red flag tokens Sophie Xhonneux David Dobre Mehrnaz Mohfakhami Leo Schwinn Gauthier Gidel 33 1 0 22 Feb 2025
Activation Steering in Neural Theorem Provers Shashank Kirtania LLMSV 51 0 0 21 Feb 2025
Understanding and Rectifying Safety Perception Distortion in VLMs Xiaohan Zou Jian Kang George Kesidis Lu Lin 84 0 0 18 Feb 2025
Jailbreaking to Jailbreak Jeremy Kritz Vaughn Robinson Robert Vacareanu Bijan Varjavand Michael Choi Bobby Gogov Scale Red Team Summer Yue Willow Primack Zifan Wang 85 0 0 09 Feb 2025
Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities Zora Che Stephen Casper Robert Kirk Anirudh Satheesh Stewart Slocum ... Zikui Cai Bilal Chughtai Y. Gal Furong Huang Dylan Hadfield-Menell MU AAML ELM 58 2 0 03 Feb 2025
ICLR: In-Context Learning of Representations Core Francisco Park Andrew Lee Ekdeep Singh Lubana Yongyi Yang Maya Okawa Kento Nishi Martin Wattenberg Hidenori Tanaka AIFin 105 3 0 29 Dec 2024
SATA: A Paradigm for LLM Jailbreak via Simple Assistive Task Linkage Xiaoning Dong Wenbo Hu Wei Xu Tianxing He 67 0 0 19 Dec 2024
All or None: Identifiable Linear Properties of Next-token Predictors in Language Modeling Emanuele Marconato Sébastien Lachapelle Sebastian Weichwald Luigi Gresele 39 3 0 30 Oct 2024
Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks Samuele Poppi Zheng-Xin Yong Yifei He Bobbie Chern Han Zhao Aobo Yang Jianfeng Chi AAML 31 11 0 23 Oct 2024
On the Role of Attention Heads in Large Language Model Safety Z. Zhou Haiyang Yu Xinghua Zhang Rongwu Xu Fei Huang Kun Wang Yang Liu Junfeng Fang Yongbin Li 27 5 0 17 Oct 2024
Improving Instruction-Following in Language Models through Activation Steering Alessandro Stolfo Vidhisha Balachandran Safoora Yousefi Eric Horvitz Besmira Nushi LLMSV 40 13 0 15 Oct 2024
Towards Interpreting Visual Information Processing in Vision-Language Models Clement Neo Luke Ong Philip H. S. Torr Mor Geva David M. Krueger Fazl Barez 78 6 0 09 Oct 2024
Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation Xinpeng Wang Chengzhi Hu Paul Röttger Barbara Plank 33 5 0 04 Oct 2024
Robust LLM safeguarding via refusal feature adversarial training L. Yu Virginie Do Karen Hambardzumyan Nicola Cancedda AAML 31 9 0 30 Sep 2024
An Adversarial Perspective on Machine Unlearning for AI Safety Jakub Łucki Boyi Wei Yangsibo Huang Peter Henderson F. Tramèr Javier Rando MU AAML 51 31 0 26 Sep 2024
A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models Daking Rai Yilun Zhou Shi Feng Abulhair Saparov Ziyu Yao 47 18 0 02 Jul 2024
Gemma: Open Models Based on Gemini Research and Technology Gemma Team Gemma Team Thomas Mesnard Cassidy Hardin Robert Dadashi Surya Bhupatiraju ... Armand Joulin Noah Fiedel Evan Senter Alek Andreev Kathleen Kenealy VLM LLMAG 120 415 0 13 Mar 2024
Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications Boyi Wei Kaixuan Huang Yangsibo Huang Tinghao Xie Xiangyu Qi Mengzhou Xia Prateek Mittal Mengdi Wang Peter Henderson AAML 49 78 0 07 Feb 2024
A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity Andrew Lee Xiaoyan Bai Itamar Pres Martin Wattenberg Jonathan K. Kummerfeld Rada Mihalcea 46 95 0 03 Jan 2024
The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets Samuel Marks Max Tegmark HILM 83 164 0 10 Oct 2023
Generative Agents: Interactive Simulacra of Human Behavior J. Park Joseph C. O'Brien Carrie J. Cai Meredith Ringel Morris Percy Liang Michael S. Bernstein LM&Ro AI4CE 204 1,701 0 07 Apr 2023
Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations Atticus Geiger Zhengxuan Wu Christopher Potts Thomas F. Icard Noah D. Goodman CML 73 98 0 05 Mar 2023
Toy Models of Superposition Nelson Elhage Tristan Hume Catherine Olsson Nicholas Schiefer T. Henighan ... Sam McCandlish Jared Kaplan Dario Amodei Martin Wattenberg C. Olah AAML MILM 117 183 0 21 Sep 2022
Training language models to follow instructions with human feedback Long Ouyang Jeff Wu Xu Jiang Diogo Almeida Carroll L. Wainwright ... Amanda Askell Peter Welinder Paul Christiano Jan Leike Ryan J. Lowe OSLM ALM 301 11,730 0 04 Mar 2022
The Pile: An 800GB Dataset of Diverse Text for Language Modeling Leo Gao Stella Biderman Sid Black Laurence Golding Travis Hoppe ... Horace He Anish Thite Noa Nabeshima Shawn Presser Connor Leahy AIMat 236 1,508 0 31 Dec 2020