English
Related papers

Related papers: Detection Is Cheap, Routing Is Learned: Why Refusa…

200 papers

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

Coherent discourse is distinguished from a mere collection of utterances by the satisfaction of a diverse set of constraints, for example choice of expression, logical relation between denoted events, and implicit compatibility with…

Computation and Language · Computer Science 2021-05-11 Anne Beyer , Sharid Loáiciga , David Schlangen

Refusal on harmful prompts is a key safety behaviour in instruction-tuned large language models (LLMs), yet the internal causes of this behaviour remain poorly understood. We study two public instruction-tuned models, Gemma-2-2B-IT and…

Computation and Language · Computer Science 2026-04-29 Nirmalendu Prakash , Yeo Wei Jie , Amir Abdullah , Ranjan Satapathy , Erik Cambria , Roy Ka Wei Lee

Current evaluation of mathematical reasoning in language models relies primarily on answer accuracy, potentially masking fundamental failures in logical computation. We introduce a diagnostic framework that distinguishes genuine…

Computation and Language · Computer Science 2025-12-02 Subramanyam Sahoo , Vinija Jain , Saanidhya Vats , Siddharth Mohapatra , Rui Min , Aman Chadha , Divya Chaudhary

Large language models are increasingly used to mediate everyday interpersonal dilemmas, yet how their advisory defaults interact with the concentrated moral orders of specific communities remains poorly understood. This article compares…

Computers and Society · Computer Science 2026-04-27 Tom van Nuenen

Humans do not just find mistakes after the fact -- we often catch them mid-stream because 'reflection' is tied to the goal and its constraints. Today's large language models produce reasoning tokens and 'reflective' text, but is it…

Artificial Intelligence · Computer Science 2025-10-24 Sion Weatherhead , Flora Salim , Aaron Belbasis

Training a language model to be both helpful and harmless requires careful calibration of refusal behaviours: Models should refuse to follow malicious instructions or give harmful advice (e.g."how do I kill someone?"), but they should not…

Computation and Language · Computer Science 2025-03-05 Xinpeng Wang , Chengzhi Hu , Paul Röttger , Barbara Plank

Safety alignment in large language models (LLMs), particularly for cybersecurity tasks, primarily focuses on preventing misuse. While this approach reduces direct harm, it obscures a complementary failure mode: denial of assistance to…

Cryptography and Security · Computer Science 2026-03-12 David Campbell , Neil Kale , Udari Madhushani Sehwag , Bert Herring , Nick Price , Dan Borges , Alex Levinson , Christina Q Knight

With the rapid advancement of Vision Language Models (VLMs), refusal mechanisms have become a critical component for ensuring responsible and safe model behavior. However, existing refusal strategies are largely \textit{one-size-fits-all}…

Computer Vision and Pattern Recognition · Computer Science 2026-02-10 Jiaxi Yang , Shicheng Liu , Yuchen Yang , Dongwon Lee

Language models are commonly fine-tuned for safety alignment to refuse harmful prompts. One approach fine-tunes them to generate categorical refusal tokens that distinguish different refusal types before responding. In this work, we…

Artificial Intelligence · Computer Science 2026-03-17 Rishab Alagharu , Ishneet Sukhvinder Singh , Shaibi Shamsudeen , Zhen Wu , Ashwinee Panda

Refusal behavior in aligned LLMs is often viewed as model-specific, yet we hypothesize it stems from a universal, low-dimensional semantic circuit shared across models. To test this, we introduce Trajectory Replay via Concept-Basis…

Computation and Language · Computer Science 2026-01-27 Tony Cristofano

When a user tells an AI system that someone "should not" take an action, the system ought to treat this as a prohibition. Yet many large language models do the opposite: they interpret negated instructions as affirmations. We audited 16…

Artificial Intelligence · Computer Science 2026-01-30 Katherine Elkins , Jon Chun

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

Online sexism increasingly appears in subtle, context-dependent forms that evade traditional detection methods. Its interpretation often depends on overlapping linguistic, psychological, legal, and cultural dimensions, which produce mixed…

Computation and Language · Computer Science 2026-01-08 Anwar Alajmi , Gabriele Pergola

Reward-model-based fine-tuning is a central paradigm in aligning Large Language Models with human preferences. However, such approaches critically rely on the assumption that proxy reward models accurately reflect intended supervision, a…

Computation and Language · Computer Science 2026-01-21 Zixuan Liu , Siavash H. Khajavi , Guangkai Jiang , Xinru Liu

Most of previous machine learning algorithms are proposed based on the i.i.d. hypothesis. However, this ideal assumption is often violated in real applications, where selection bias may arise between training and testing process. Moreover,…

Computer Vision and Pattern Recognition · Computer Science 2018-08-24 Zheyan Shen , Peng Cui , Kun Kuang , Bo Li , Peixuan Chen

Alignment faking (AF) refers to a model strategically complying with a training objective to avoid behavioural modification while preserving its deployment preferences. Understanding when and why AF arises matters as models grow better at…

Artificial Intelligence · Computer Science 2026-05-28 Nathaniel Mitrani Hadida , Rhea Karty , David Williams-King , Alan Cooney

Learning with rejection has been a prototypical model for studying the human-AI interaction on prediction tasks. Upon the arrival of a sample instance, the model first uses a rejector to decide whether to accept and use the AI predictor to…

Machine Learning · Computer Science 2024-04-23 Xiaocheng Li , Shang Liu , Chunlin Sun , Hanzhao Wang

An agent must act on the situation before it, learn what it cannot yet represent, and model other agents well enough to coordinate. These faculties are usually realized by separate mechanisms, yet they share a failure mode: the situation…

Neurons and Cognition · Quantitative Biology 2026-05-26 Chainarong Amornbunchornvej

This manuscript extends our previous multimodal human-robot interaction system by introducing a controlled ablation study of the three modules that most strongly influence end-to-end performance: the large language model used for action…

Robotics · Computer Science 2026-05-05 Zi Tian , Guanting Shen
‹ Prev 1 3 4 5 6 7 10 Next ›