English
Related papers

Related papers: Why Do Some Language Models Fake Alignment While O…

200 papers

The development of large language models leads to the formation of a pre-train-then-align paradigm, in which the model is typically pre-trained on a large text corpus and undergoes a tuning stage to align the model with human preference or…

Computation and Language · Computer Science 2025-03-19 Kaiser Sun , Mark Dredze

An LLM's factuality and refusal training can be compromised by simple changes to a prompt. Models often adopt user beliefs (sycophancy) or satisfy inappropriate requests which are wrapped within special text (jailbreaking). We explore…

Machine Learning · Computer Science 2025-11-03 Alex Irpan , Alexander Matt Turner , Mark Kurzeja , David K. Elson , Rohin Shah

Small and mid-sized generative language models have gained increasing attention. Their size and availability make them amenable to being analyzed at a behavioral as well as a representational level, allowing investigations of how these…

Machine Learning · Computer Science 2025-04-11 Lorenz Linhardt , Tom Neuhäuser , Lenka Tětková , Oliver Eberle

Large Language Models (LLMs) are rarely static and are frequently updated in practice. A growing body of alignment research has shown that models initially deemed "aligned" can exhibit misaligned behavior after fine-tuning, such as…

Machine Learning · Computer Science 2026-02-02 Yavuz Bakman , Duygu Nur Yaldiz , Salman Avestimehr , Sai Praneeth Karimireddy

This study evaluates Direct Preference Optimization (DPO) and its variants for aligning Large Language Models (LLMs) with human preferences, testing three configurations: (1) with Supervised Fine Tuning (SFT), (2) without SFT, and (3)…

Computation and Language · Computer Science 2025-02-11 Amir Saeidi , Shivanshu Verma , Md Nayem Uddin , Chitta Baral

Large language models have been shown to suffer from reasoning inconsistency issues. That is, they fail more in situations unfamiliar to the training data, even though exact or very similar reasoning paths exist in more common cases that…

Computation and Language · Computer Science 2025-02-04 Ben Zhou , Sarthak Jain , Yi Zhang , Qiang Ning , Shuai Wang , Yassine Benajiba , Dan Roth

Instruction tuning has been widely adopted to ensure large language models (LLMs) follow user instructions effectively. The resulting instruction-following capabilities of LLMs heavily rely on the instruction datasets used for tuning.…

Artificial Intelligence · Computer Science 2025-02-27 Zhangchen Xu , Fengqing Jiang , Luyao Niu , Bill Yuchen Lin , Radha Poovendran

For Large Language Models (LLMs) to be reliably deployed, models must effectively know when not to answer: abstain. Reasoning models, in particular, have gained attention for impressive performance on complex tasks. However, reasoning…

Artificial Intelligence · Computer Science 2026-04-03 Abinitha Gourabathina , Inkit Padhi , Manish Nagireddy , Subhajit Chaudhury , Prasanna Sattigeri

A well-calibrated model should express confidence that matches its actual accuracy -- when it claims 80\% confidence, it should be correct 80\% of the time. While large language models (LLMs) have achieved remarkable performance across…

Artificial Intelligence · Computer Science 2025-12-19 Lukas Nel

Safety alignment can make frontier LMs overly conservative, degrading collaboration via hedging or false refusals. We present a lightweight toolkit with three parts: (1) Victor Calibration (VC), a multi-pass protocol that elicits a scalar…

Software Engineering · Computer Science 2025-12-30 Victor Stasiuc

Large language models (LLMs) are increasingly embedded in healthcare workflows for documentation, education, and clinical decision support. However, these systems are trained on large text corpora that encode existing biases, including sex…

Computation and Language · Computer Science 2026-02-05 Isabel Tsintsiper , Sheng Wong , Beth Albert , Shaun P Brennecke , Gabriel Davis Jones

Large Language Models (LLMs) are widely used for automated code generation, yet their apparent successes often mask a tension between pretraining objectives and alignment choices. While pretraining encourages models to exploit all available…

Software Engineering · Computer Science 2025-12-25 Oussama Ben Sghaier , Kevin Delcourt , Houari Sahraoui

Reasoning models frequently agree with incorrect user suggestions -- a behavior known as sycophancy. However, it is unclear where in the reasoning trace this agreement originates and how strong the commitment is. We introduce…

Artificial Intelligence · Computer Science 2026-02-10 Jacek Duszenko

Safety alignment is a key requirement for building reliable Artificial General Intelligence. Despite significant advances in safety alignment, we observe that minor latent shifts can still trigger unsafe responses in aligned models. We…

Machine Learning · Computer Science 2025-06-23 Tianle Gu , Kexin Huang , Zongqi Wang , Yixu Wang , Jie Li , Yuanqi Yao , Yang Yao , Yujiu Yang , Yan Teng , Yingchun Wang

Optimizing large language models (LLMs) for downstream use cases often involves the customization of pre-trained LLMs through further fine-tuning. Meta's open release of Llama models and OpenAI's APIs for fine-tuning GPT-3.5 Turbo on custom…

Computation and Language · Computer Science 2023-10-06 Xiangyu Qi , Yi Zeng , Tinghao Xie , Pin-Yu Chen , Ruoxi Jia , Prateek Mittal , Peter Henderson

Large language models (LLMs) provide a compelling foundation for building generally-capable AI agents. These agents may soon be deployed at scale in the real world, representing the interests of individual humans (e.g., AI assistants) or…

Multiagent Systems · Computer Science 2024-12-16 Aron Vallinder , Edward Hughes

LLM self-explanations are often presented as a promising tool for AI oversight, yet their faithfulness to the model's true reasoning process is poorly understood. Existing faithfulness metrics have critical limitations, typically relying on…

Artificial Intelligence · Computer Science 2026-02-04 Harry Mayne , Justin Singh Kang , Dewi Gould , Kannan Ramchandran , Adam Mahdi , Noah Y. Siegel

Language models are trained to follow instructions, but they are also powerful pattern completers. What happens when these two objectives conflict? We construct conversations in which a user instruction to behave in a target way T (e.g.,…

Computation and Language · Computer Science 2026-05-21 Carolina Camassa , Derek Shiller

Large language models (LLMs) increasingly operate in multi-agent and safety-critical settings, raising open questions about how their vulnerabilities scale when models interact adversarially. This study examines whether larger models can…

Machine Learning · Computer Science 2026-01-05 Samuel Nathanson , Rebecca Williams , Cynthia Matuszek

This report examines whether advanced AIs that perform well in training will be doing so in order to gain power later -- a behavior I call "scheming" (also sometimes called "deceptive alignment"). I conclude that scheming is a disturbingly…

Computers and Society · Computer Science 2023-11-29 Joe Carlsmith