English
Related papers

Related papers: Refusal in LLMs is an Affine Function

200 papers

Model editing aims at selectively updating a small subset of a neural model's parameters with an interpretable strategy to achieve desired modifications. It can significantly reduce computational costs to adapt to large language models…

Computation and Language · Computer Science 2025-03-20 Shichen Li , Zhongqing Wang , Zheyu Zhao , Yue Zhang , Peifeng Li

Large Language Models (LLMs) rely on safety alignment to obey safe requests while refusing harmful ones. However, traditional refusal mechanisms often lead to "rigid rejection," where a general template (e.g., "I cannot fulfill this…

Computation and Language · Computer Science 2026-05-11 Ying Zhang , Congyu Qiao , Xin Geng , Ning Xu

Activation steering is a promising technique for controlling LLM behavior by adding semantically meaningful vectors directly into a model's hidden states during inference. It is often framed as a precise, interpretable, and potentially…

Machine Learning · Computer Science 2026-02-17 Anton Korznikov , Andrey Galichin , Alexey Dontsov , Oleg Y. Rogov , Ivan Oseledets , Elena Tutubalina

Large language models can resist task-misaligned activation steering during inference, sometimes recovering mid-generation to produce improved responses even when steering remains active. We term this Endogenous Steering Resistance (ESR).…

With the introduction of (large) language models, there has been significant concern about the unintended bias such models may inherit from their training data. A number of studies have shown that such models propagate gender stereotypes,…

Computation and Language · Computer Science 2024-08-20 Rameez Qureshi , Naïm Es-Sebbani , Luis Galárraga , Yvette Graham , Miguel Couceiro , Zied Bouraoui

This study reveals a previously unexplored vulnerability in the safety alignment of Large Language Models (LLMs). Existing aligned LLMs predominantly respond to unsafe queries with refusals, which often begin with a fixed set of prefixes…

Cryptography and Security · Computer Science 2026-01-28 Yangyang Guo , Ziwei Xu , Si Liu , Zhiming Zheng , Mohan Kankanhalli

The growing prominence of LLMs has led to an increase in the development of AI tutoring systems. These systems are crucial in providing underrepresented populations with improved access to valuable education. One important area of education…

Computation and Language · Computer Science 2024-10-03 Ryan Shea , Aymen Kallala , Xin Lucy Liu , Michael W. Morris , Zhou Yu

Large Language Models (LLMs) have demonstrated impressive capabilities in complex reasoning tasks. However, they can be easily misled by unfaithful arguments during conversations, even when their original statements are correct. To this…

Computation and Language · Computer Science 2025-01-03 Yong Zhao , Yang Deng , See-Kiong Ng , Tat-Seng Chua

The capabilities of Large Language Models (LLMs) are routinely evaluated by other LLMs trained to predict human preferences. This framework--known as LLM-as-a-judge--is highly scalable and relatively low cost. However, it is also vulnerable…

Computation and Language · Computer Science 2026-02-03 Lisa Alazraki , Tan Yi-Chern , Jon Ander Campos , Maximilian Mozes , Marek Rei , Max Bartolo

Recent advancements in Large Language Models (LLMs) and Reinforcement Learning (RL) have shown significant promise in decision-making tasks. Nevertheless, for large-scale industrial decision problems, both approaches face distinct…

Artificial Intelligence · Computer Science 2025-06-04 Xu Wan , Wenyue Xu , Chao Yang , Mingyang Sun

Vision-language-action (VLA) models provide a promising paradigm for scalable robotic manipulation, yet their reliance on success-only behavioral cloning leaves them brittle; lacking corrective training signals, minor execution errors…

Fine-tuning large language models (LLMs) to adapt to evolving safety policies is costly and impractical. Mechanistic interpretability enables inference-time control through latent activation steering, yet its potential for precise,…

Machine Learning · Computer Science 2025-06-06 Shaona Ghosh , Amrita Bhattacharjee , Yftah Ziser , Christopher Parisien

Current alignment evaluation mostly measures whether models encode dangerous concepts and whether they refuse harmful requests. Both miss the layer where alignment often operates: routing from concept detection to behavioral policy. We…

Machine Learning · Computer Science 2026-05-04 Gregory N. Frank

Large Language Models (LLMs) demonstrate increasing conversational fluency, yet instilling them with nuanced, human-like emotional expression remains a significant challenge. Current alignment techniques often address surface-level output…

Computation and Language · Computer Science 2025-11-25 Niranjan Chebrolu , Gerard Christopher Yeo , Kokil Jaidka

Large language models excel at complex instructions yet struggle to deviate from their helpful assistant persona, as post-training instills strong priors that resist conflicting instructions. We introduce system prompt strength, a…

Computation and Language · Computer Science 2026-01-13 Yijiang River Dong , Tiancheng Hu , Zheng Hui , Nigel Collier

Unfair stereotypical biases (e.g., gender, racial, or religious biases) encoded in modern pretrained language models (PLMs) have negative ethical implications for widespread adoption of state-of-the-art language technology. To remedy for…

Computation and Language · Computer Science 2021-09-09 Anne Lauscher , Tobias Lüken , Goran Glavaš

LLMs increasingly exhibit over-refusal behavior, where safety mechanisms cause models to reject benign instructions that seemingly resemble harmful content. This phenomenon diminishes utility in production applications that repeatedly rely…

Computation and Language · Computer Science 2026-04-21 Utsav Maskey , Sumit Yadav , Mark Dras , Usman Naseem

Controlling specific behaviors in large language models while preserving their general capabilities is a central challenge for safe and reliable artificial intelligence deployment. Current steering methods, such as vector addition and…

Machine Learning · Computer Science 2025-10-31 Hieu M. Vu , Tan M. Nguyen

Large language models (LLMs) can sometimes detect when they are being evaluated and adjust their behavior to appear more aligned, compromising the reliability of safety evaluations. In this paper, we show that adding a steering vector to an…

Computation and Language · Computer Science 2026-03-03 Tim Tian Hua , Andrew Qin , Samuel Marks , Neel Nanda

Activation engineering is becoming increasingly popular as a means of online control of large language models (LLMs). In this work, we extend the idea of inference-time steering with vectors that represent a behavioral direction of interest…

Machine Learning · Computer Science 2024-11-26 Christopher M. Ackerman