English
Related papers

Related papers: Emergently Misaligned Language Models Show Behavio…

200 papers

Fine-tuning Large Language Models (LLMs) on benign narrow data can sometimes induce broad harmful behaviors, a vulnerability termed emergent misalignment (EM). While prior work links these failures to specific directions in the activation…

Computation and Language · Computer Science 2026-05-12 Krishak Aneja , Manas Mittal , Anmol Goel , Ponnurangam Kumaraguru , Vamshi Krishna Bonagiri

Emergent misalignment, where fine-tuning on narrow, non-harmful tasks induces harmful behaviors, poses a key challenge for AI safety in LLMs. Despite growing empirical evidence, its underlying mechanism remains unclear. To uncover the…

Artificial Intelligence · Computer Science 2026-05-05 Gouki Minegishi , Hiroki Furuta , Takeshi Kojima , Yusuke Iwasawa , Yutaka Matsuo

Fine-tuning large language models on narrow data with harmful content produces broadly misaligned behavior on unrelated prompts, a phenomenon known as emergent misalignment. We propose that emergent misalignment involves persona-model…

Computation and Language · Computer Science 2026-05-26 Davi Bastos Costa , Renato Vicente

Large language models (LLMs) are increasingly used in decision-making contexts, but when they present answers without signaling low confidence, users may unknowingly act on erroneous outputs. Prior work shows that LLMs maintain internal…

Computation and Language · Computer Science 2025-10-23 Mark Steyvers , Catarina Belem , Padhraic Smyth

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

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

Large language models (LLMs) are increasingly employed in information-seeking and decision-making tasks. Despite their broad utility, LLMs tend to generate information that conflicts with real-world facts, and their persuasive style can…

Computation and Language · Computer Science 2024-09-19 Arslan Chaudhry , Sridhar Thiagarajan , Dilan Gorur

Large Language Models (LLMs) can acquire deceptive behaviors through backdoor attacks, where the model executes prohibited actions whenever secret triggers appear in the input. Existing safety training methods largely fail to address this…

Cryptography and Security · Computer Science 2025-10-08 Guangyu Shen , Siyuan Cheng , Xiangzhe Xu , Yuan Zhou , Hanxi Guo , Zhuo Zhang , Xiangyu Zhang

Large language models (LLMs) often produce confident yet incorrect answers, which can lead to risky failures in real-world applications. We study whether post-training can make a model's self-assessment explicit: when the model is…

Machine Learning · Computer Science 2026-05-15 Junyu Guo , Shangding Gu , Ming Jin , Costas Spanos , Javad Lavaei

While brain-aligned large language models (LLMs) have garnered attention for their potential as cognitive models and for potential for enhanced safety and trustworthiness in AI, the role of this brain alignment for linguistic competence…

Computation and Language · Computer Science 2026-03-25 Gabriele Merlin , Mariya Toneva

Large Language Models (LLMs) are leading a new technological revolution as one of the most promising research streams toward artificial general intelligence. The scaling of these models, accomplished by increasing the number of parameters…

Machine Learning · Computer Science 2025-03-17 Leonardo Berti , Flavio Giorgi , Gjergji Kasneci

Large Language Models (LLMs) have made significant advances in natural language processing, but their underlying mechanisms are often misunderstood. Despite exhibiting coherent answers and apparent reasoning behaviors, LLMs rely on…

Computation and Language · Computer Science 2024-08-05 Bo Zhou , Daniel Geißler , Paul Lukowicz

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

Emergent misalignment (EM), where fine-tuning on a narrow task (like insecure code) causes broad misalignment across unrelated domains, was first demonstrated by Betley et al. (2025). We conduct the most comprehensive EM study to date,…

Machine Learning · Computer Science 2026-05-13 Joel Schreiber , Ariel Goldstein

Large language models (LLMs) have demonstrated revolutionary capabilities in understanding complex contexts and performing a wide range of tasks. However, LLMs can also answer questions that are unethical or harmful, raising concerns about…

Cryptography and Security · Computer Science 2025-04-15 Kang Yang , Guanhong Tao , Xun Chen , Jun Xu

Recent work has discovered that large language models can develop broadly misaligned behaviors after being fine-tuned on narrowly harmful datasets, a phenomenon known as emergent misalignment (EM). However, the fundamental mechanisms…

Machine Learning · Computer Science 2025-11-05 Daniel Aarao Reis Arturi , Eric Zhang , Andrew Ansah , Kevin Zhu , Ashwinee Panda , Aishwarya Balwani

Large language models (LLMs) may exhibit unintended or undesirable behaviors. Recent works have concentrated on aligning LLMs to mitigate harmful outputs. Despite these efforts, some anomalies indicate that even a well-conducted alignment…

Computation and Language · Computer Science 2025-09-24 Jiaming Ji , Kaile Wang , Tianyi Qiu , Boyuan Chen , Jiayi Zhou , Changye Li , Hantao Lou , Juntao Dai , Yunhuai Liu , Yaodong Yang

Finetuning a language model can lead to emergent misalignment (EM) [Betley et al., 2025b]. Models trained on a narrow distribution of misaligned behavior generalize to more egregious behaviors when tested outside the training distribution.…

Machine Learning · Computer Science 2026-04-29 Jan Dubiński , Jan Betley , Anna Sztyber-Betley , Daniel Tan , Owain Evans

This paper investigates the impact of incorrect data on the performance and safety of large language models (LLMs), specifically gpt-4o, during supervised fine-tuning (SFT). Although LLMs become increasingly vital across broad domains like…

Computation and Language · Computer Science 2025-09-25 Jian Ouyang , Arman T , Ge Jin

A fundamental open challenge in modern LLM scaling is the lack of understanding around emergent capabilities. In particular, language model pretraining loss is known to be highly predictable as a function of compute. However, downstream…

Machine Learning · Computer Science 2024-11-26 Charlie Snell , Eric Wallace , Dan Klein , Sergey Levine