English
Related papers

Related papers: The Missing Half: Unveiling Training-time Implicit…

200 papers

Backdoor attacks inject poisoning samples during training, with the goal of forcing a machine learning model to output an attacker-chosen class when presented a specific trigger at test time. Although backdoor attacks have been demonstrated…

As machine learning models become more capable, they have exhibited increased potential in solving complex tasks. One of the most promising directions uses deep reinforcement learning to train autonomous agents in computer network defense…

Machine Learning · Computer Science 2023-10-23 Elizabeth Bates , Vasilios Mavroudis , Chris Hicks

This paper presents a comprehensive empirical study on the safety alignment capabilities. We evaluate what matters for safety alignment in LLMs and LRMs to provide essential insights for developing more secure and reliable AI systems. We…

Computation and Language · Computer Science 2026-02-25 Xing Li , Hui-Ling Zhen , Lihao Yin , Xianzhi Yu , Zhenhua Dong , Mingxuan Yuan

Safety fine-tuning of language models typically requires a curated adversarial dataset. We take a different approach: score each candidate prompt's difficulty by how often the target model's own rollouts are judged harmful, then fine-tune…

Machine Learning · Computer Science 2026-05-06 Prakhar Gupta , Garv Shah , Donghua Zhang

LLMs are trained to refuse harmful instructions, but do they truly understand harmfulness beyond just refusing? Prior work has shown that LLMs' refusal behaviors can be mediated by a one-dimensional subspace, i.e., a refusal direction. In…

Computation and Language · Computer Science 2025-12-16 Jiachen Zhao , Jing Huang , Zhengxuan Wu , David Bau , Weiyan Shi

The desire to use reinforcement learning in safety-critical settings has inspired a recent interest in formal methods for learning algorithms. Existing formal methods for learning and optimization primarily consider the problem of…

Artificial Intelligence · Computer Science 2019-06-05 Nathan Fulton , Andre Platzer

We study a security threat to reinforcement learning where an attacker poisons the learning environment to force the agent into executing a target policy chosen by the attacker. As a victim, we consider RL agents whose objective is to find…

Machine Learning · Computer Science 2020-08-20 Amin Rakhsha , Goran Radanovic , Rati Devidze , Xiaojin Zhu , Adish Singla

Large language models (LLMs) are now deployed at unprecedented scale, assisting millions of users in daily tasks. However, the risk of these models assisting unlawful activities remains underexplored. In this study, we define this high-risk…

Computers and Society · Computer Science 2025-11-27 Xing Wang , Huiyuan Xie , Yiyan Wang , Chaojun Xiao , Huimin Chen , Holli Sargeant , Felix Steffek , Jie Shao , Zhiyuan Liu , Maosong Sun

Backdoor attacks pose a significant threat to the integrity and reliability of Artificial Intelligence (AI) models, enabling adversaries to manipulate model behavior by injecting poisoned data with hidden triggers. These attacks can lead to…

Machine Learning · Computer Science 2026-03-31 Osama Wehbi , Sarhad Arisdakessian , Omar Abdel Wahab , Azzam Mourad , Hadi Otrok , Jamal Bentahar

LLM-based coding agents are rapidly being deployed in software development, yet their safety implications remain poorly understood. These agents, while capable of accelerating software development, may exhibit unsafe behaviors during normal…

Artificial Intelligence · Computer Science 2025-08-26 Matous Kozak , Roshanak Zilouchian Moghaddam , Siva Sivaraman

Safety-aligned LLMs go through refusal training to reject harmful requests, but whether these mechanisms remain effective under emotionally charged stimuli is unexplored. We introduce FreakOut-LLM, a framework investigating whether…

Cryptography and Security · Computer Science 2026-04-08 Daniel Kuznetsov , Ofir Cohen , Karin Shistik , Rami Puzis , Asaf Shabtai

Construction remains one of the most hazardous sectors. Recent advancements in AI, particularly Large Language Models (LLMs), offer promising opportunities for enhancing workplace safety. However, responsible integration of LLMs requires…

Artificial Intelligence · Computer Science 2024-11-14 Farouq Sammour , Jia Xu , Xi Wang , Mo Hu , Zhenyu Zhang

LLM safety evaluations predominantly test models in isolation, yet deployed AI agents increasingly operate within persistent social environments alongside other agents. We introduce a Moltbook-style simulation platform where thousands of…

Artificial Intelligence · Computer Science 2026-05-28 Aman Priyanshu , Supriti Vijay , Esha Pahwa

Safety fine-tuning helps align Large Language Models (LLMs) with human preferences for their safe deployment. To better understand the underlying factors that make models safe via safety fine-tuning, we design a synthetic data generation…

Machine Learning · Computer Science 2024-08-22 Samyak Jain , Ekdeep Singh Lubana , Kemal Oksuz , Tom Joy , Philip H. S. Torr , Amartya Sanyal , Puneet K. Dokania

Large Language Models (LLMs) have demonstrated impressive capabilities in natural language tasks, but their safety and morality remain contentious due to their training on internet text corpora. To address these concerns, alignment…

Computation and Language · Computer Science 2024-08-06 Mohammad Bahrami Karkevandi , Nishant Vishwamitra , Peyman Najafirad

We systematically evaluate the quality of widely used adversarial safety datasets from two perspectives: in isolation and in practice. In isolation, we examine how well these datasets reflect real-world adversarial attacks based on three…

Cryptography and Security · Computer Science 2026-04-24 Shahriar Golchin , Marc Wetter

As LLMs become more widely deployed, there is increasing interest in directly optimizing for feedback from end users (e.g. thumbs up) in addition to feedback from paid annotators. However, training to maximize human feedback creates a…

Machine Learning · Computer Science 2025-02-25 Marcus Williams , Micah Carroll , Adhyyan Narang , Constantin Weisser , Brendan Murphy , Anca Dragan

Large language models exhibit systematic vulnerabilities to adversarial attacks despite extensive safety alignment. We provide a mechanistic analysis revealing that position-dependent gradient weakening during autoregressive training…

Machine Learning · Computer Science 2025-11-18 Thong Bach , Dung Nguyen , Thao Minh Le , Truyen Tran

Large language models (LLMs) are foundational explorations to artificial general intelligence, yet their alignment with human values via instruction tuning and preference learning achieves only superficial compliance. Here, we demonstrate…

Computation and Language · Computer Science 2025-06-04 Jiawei Lian , Jianhong Pan , Lefan Wang , Yi Wang , Shaohui Mei , Lap-Pui Chau

Large Language Models (LLMs) rapidly reshape modern life, advancing fields from healthcare to education and beyond. However, alongside their remarkable capabilities lies a significant threat: the susceptibility of these models to…

Computation and Language · Computer Science 2025-05-16 Michael Fire , Yitzhak Elbazis , Adi Wasenstein , Lior Rokach