English
Related papers

Related papers: Understanding Emergent Misalignment via Feature Su…

200 papers

Large language models (LLMs) undergo alignment training to avoid harmful behaviors, yet the resulting safeguards remain brittle: jailbreaks routinely bypass them, and fine-tuning on narrow domains can induce ``emergent misalignment'' that…

Computation and Language · Computer Science 2026-04-13 Hadas Orgad , Boyi Wei , Kaden Zheng , Martin Wattenberg , Peter Henderson , Seraphina Goldfarb-Tarrant , Yonatan Belinkov

The deployment of large language models (LLMs) raises significant ethical and safety concerns. While LLM alignment techniques are adopted to improve model safety and trustworthiness, adversaries can exploit these techniques to undermine…

Cryptography and Security · Computer Science 2026-04-10 Rui Zhang , Hongwei Li , Yun Shen , Xinyue Shen , Wenbo Jiang , Guowen Xu , Yang Liu , Michael Backes , Yang Zhang

Lifelong multimodal agents must continuously adapt to new tasks through post-training, but this creates a fundamental tension between acquiring capabilities and preserving safety alignment. We demonstrate that fine-tuning aligned…

Artificial Intelligence · Computer Science 2026-03-17 Idhant Gulati , Shivam Raval

Fine-tuning LLMs on narrowly harmful datasets can lead to behavior that is broadly misaligned with respect to human values. To understand when and how this emergent misalignment occurs, we develop a comprehensive framework for detecting and…

Machine Learning · Computer Science 2025-08-28 Julian Arnold , Niels Lörch

Fine-tuning language models on narrowly harmful data causes emergent misalignment (EM) -- behavioral failures extending far beyond training distributions. Recent work demonstrates compartmentalization of misalignment behind contextual…

Computation and Language · Computer Science 2026-03-06 Rohan Saxena

Fine-tuning Large Language Models (LLMs) for downstream tasks often compromises safety alignment, even when using parameter-efficient methods like LoRA. In this work, we uncover a notable property: fine-tuned models preserve the geometric…

Machine Learning · Computer Science 2025-11-25 Thong Bach , Thanh Nguyen-Tang , Dung Nguyen , Thao Minh Le , Truyen Tran

The alignment problem refers to concerns regarding powerful intelligences, ensuring compatibility with human preferences and values as capabilities increase. Current large language models (LLMs) show misaligned behaviors, such as strategic…

Computation and Language · Computer Science 2026-03-10 Roshni Lulla , Fiona Collins , Sanaya Parekh , Thilo Hagendorff , Jonas Kaplan

Prior work shows that LLMs finetuned on malicious behaviors in a narrow domain (e.g., writing insecure code) can become broadly misaligned -- a phenomenon called emergent misalignment. We investigate whether this extends from conventional…

Machine Learning · Computer Science 2025-07-11 James Chua , Jan Betley , Mia Taylor , Owain Evans

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. However, their tendency to exhibit sycophantic behavior - excessively agreeing with or flattering users - poses…

Computation and Language · Computer Science 2025-01-30 Lars Malmqvist

Recent studies on Neural Collapse (NC) reveal that, under class-balanced conditions, the class feature means and classifier weights spontaneously align into a simplex equiangular tight frame (ETF). In long-tailed regimes, however, severe…

Machine Learning · Computer Science 2025-12-10 Jinping Wang , Zhiqiang Gao , Zhiwu Xie

The rapid development of large language models (LLMs) has not only provided numerous opportunities but also presented significant challenges. This becomes particularly evident when LLMs inadvertently generate harmful or toxic content,…

Computation and Language · Computer Science 2024-02-20 Kai Chen , Chunwei Wang , Kuo Yang , Jianhua Han , Lanqing Hong , Fei Mi , Hang Xu , Zhengying Liu , Wenyong Huang , Zhenguo Li , Dit-Yan Yeung , Lifeng Shang , Xin Jiang , Qun Liu

Multimodal large language models (MLLMs) have exhibited remarkable performance in various visual tasks, yet still struggle with spatial reasoning. Recent efforts mitigate this by injecting geometric features from 3D foundation models, but…

Computer Vision and Pattern Recognition · Computer Science 2026-04-15 Zhaochen Liu , Limeng Qiao , Guanglu Wan , Tingting Jiang

Weight-perturbation evolution strategies (ES) can fine-tune billion-parameter language models with surprisingly small populations (e.g., $N\!\approx\!30$), contradicting classical zeroth-order curse-of-dimensionality intuition. We also…

Machine Learning · Computer Science 2026-02-03 Qiyao Liang , Jinyeop Song , Yizhou Liu , Jeff Gore , Ila Fiete , Risto Miikkulainen , Xin Qiu

Pretraining corpora contain extensive discourse about AI systems, yet the causal influence of this discourse on downstream alignment remains poorly understood. If prevailing descriptions of AI behaviour are predominantly negative, LLMs may…

Computation and Language · Computer Science 2026-02-23 Cameron Tice , Puria Radmard , Samuel Ratnam , Andy Kim , David Africa , Kyle O'Brien

Machine learning has demonstrated remarkable prediction accuracy over i.i.d data, but the accuracy often drops when tested with data from another distribution. In this paper, we aim to offer another view of this problem in a perspective…

Machine Learning · Computer Science 2022-06-20 Haohan Wang , Zeyi Huang , Hanlin Zhang , Yong Jae Lee , Eric Xing

Reinforcement Learning from Human Feedback (RLHF) is the standard method to align Large Language Models (LLMs) with human preferences. In this work, we introduce alignment tampering, a potential vulnerability where the LLM undergoing…

Artificial Intelligence · Computer Science 2026-05-27 Dongyoon Hahm , Dylan Hadfield-Menell , Kimin Lee

Large language models (LLMs) have emerged as powerful tools for addressing a wide range of general inquiries and tasks. Despite this, fine-tuning aligned LLMs on smaller, domain-specific datasets, critical to adapting them to specialized…

Artificial Intelligence · Computer Science 2025-02-04 Guanlin Li , Kangjie Chen , Shangwei Guo , Jie Zhang , Han Qiu , Chao Zhang , Guoyin Wang , Tianwei Zhang , Jiwei Li

The prevailing approach to aligning Large Language Models (LLMs) typically relies on human or AI feedback and assumes access to specific types of preference datasets. In our work, we question the efficacy of such datasets and explore…

Machine Learning · Computer Science 2024-03-19 Hao Sun

Robots have been increasingly better at doing tasks for humans by learning from their feedback, but still often suffer from model misalignment due to missing or incorrectly learned features. When the features the robot needs to learn to…

Robotics · Computer Science 2023-04-14 Inês Lourenço , Andreea Bobu , Cristian R. Rojas , Bo Wahlberg

Large Language Models (LLMs) drive current AI breakthroughs despite very little being known about their internal representations. In this work, we propose to shed the light on LLMs inner mechanisms through the lens of geometry. In…

Artificial Intelligence · Computer Science 2024-07-12 Randall Balestriero , Romain Cosentino , Sarath Shekkizhar