English
Related papers

Related papers: Simple Role Assignment is Extraordinarily Effectiv…

200 papers

Frontier AI systems are increasingly capable and deployed in high-stakes multi-agent environments. However, existing AI safety benchmarks largely evaluate single agents, leaving multi-agent risks such as coordination failure and conflict…

Artificial Intelligence · Computer Science 2026-05-25 Pepijn Cobben , Xuanqiang Angelo Huang , Thao Amelia Pham , Isabel Dahlgren , Terry Jingchen Zhang , Zhijing Jin

Agentic methods have emerged as a powerful and autonomous paradigm that enhances reasoning, collaboration, and adaptive control, enabling systems to coordinate and independently solve complex tasks. We extend this paradigm to safety…

Artificial Intelligence · Computer Science 2025-10-30 Juan Ren , Mark Dras , Usman Naseem

Large Language Models have shown impressive generative capabilities across diverse tasks, but their safety remains a critical concern. Existing post-training alignment methods, such as SFT and RLHF, reduce harmful outputs yet leave LLMs…

Cryptography and Security · Computer Science 2025-10-21 Zhengyue Zhao , Yingzi Ma , Somesh Jha , Marco Pavone , Patrick McDaniel , Chaowei Xiao

Existing training-time safety alignment techniques for large language models (LLMs) remain vulnerable to jailbreak attacks. Direct preference optimization (DPO), a widely deployed alignment method, exhibits limitations in both experimental…

Computation and Language · Computer Science 2025-10-31 Xuandong Zhao , Will Cai , Tianneng Shi , David Huang , Licong Lin , Song Mei , Dawn Song

Inductive reasoning is a core problem-solving capacity: humans can identify underlying principles from a few examples, which robustly generalize to novel scenarios. Recent work evaluates large language models (LLMs) on inductive reasoning…

Machine Learning · Computer Science 2024-06-03 Ruocheng Wang , Eric Zelikman , Gabriel Poesia , Yewen Pu , Nick Haber , Noah D. Goodman

Deploying AI-powered systems requires trustworthy models supporting effective human interactions, going beyond raw prediction accuracy. Concept bottleneck models promote trustworthiness by conditioning classification tasks on an…

State of the art reinforcement learning methods sometimes encounter unsafe situations. Identifying when these situations occur is of interest both for post-hoc analysis and during deployment, where it might be advantageous to call out to a…

Machine Learning · Computer Science 2025-05-29 Alexander Grushin , Walt Woods , Alvaro Velasquez , Simon Khan

Designing controllers that accomplish tasks while guaranteeing safety constraints remains a significant challenge. We often want an agent to perform well in a nominal task, such as environment exploration, while ensuring it can avoid unsafe…

Systems and Control · Electrical Eng. & Systems 2025-06-04 Azra Begzadić , Nikhil Uday Shinde , Sander Tonkens , Dylan Hirsch , Kaleb Ugalde , Michael C. Yip , Jorge Cortés , Sylvia Herbert

A large number of heuristics have been proposed to optimize the reinforcement fine-tuning of LLMs. However, inconsistent claims are made from time to time, making this area elusive. Reflecting on this situation, two fundamental questions…

Machine Learning · Computer Science 2026-01-22 Xiao Hu , Hong Xie , Tao Tan , Defu Lian , Jianyu Han

Recent work has shown that a model's input word embeddings can serve as effective control variables for steering its behavior toward outputs that satisfy desired properties. However, this has only been demonstrated for pretrained…

Computation and Language · Computer Science 2026-04-30 Baturay Saglam , Dionysis Kalogerias

Aligning large language models with human preferences must balance two competing goals: responding helpfully to legitimate requests and reliably refusing harmful ones. Most preference-based safety alignment methods collapse safety into a…

Computation and Language · Computer Science 2026-04-22 Tiankai Yang , Yi Nian , Xinyuan Li , Ruiyao Xu , Kaize Ding , Yue Zhao

An emerging field of sequential decision problems is safe Reinforcement Learning (RL), where the objective is to maximize the reward while obeying safety constraints. Being able to handle constraints is essential for deploying RL agents in…

Robotics · Computer Science 2023-03-08 Nick Bührer , Zhejun Zhang , Alexander Liniger , Fisher Yu , Luc Van Gool

As large language models (LLMs) are becoming more capable and widespread, the study of their failure cases is becoming increasingly important. Recent advances in standardizing, measuring, and scaling test-time compute suggest new…

Machine Learning · Computer Science 2025-06-26 Mahdi Sabbaghi , Paul Kassianik , George Pappas , Yaron Singer , Amin Karbasi , Hamed Hassani

Existing LLM-as-a-Judge systems suffer from three fundamental limitations: limited adaptivity to task- and domain-specific evaluation criteria, systematic biases driven by non-semantic cues such as position, length, format, and model…

Computation and Language · Computer Science 2026-02-09 Bo Yang , Lanfei Feng , Yunkui Chen , Yu Zhang , Xiao Xu , Shijian Li

Current safety alignment of foundation models largely follows a \emph{one-size-fits-all} paradigm, applying the same refusal policy across users and contexts. As a result, models may refuse requests that are unsafe for general users but…

Artificial Intelligence · Computer Science 2026-05-26 Qitao Tan , Xiaoying Song , Arman Akbari , Arash Akbari , Yanzhi Wang , Xiaoming Zhai , Lingzi Hong , Zhen Xiang , Jin Lu , Geng Yuan

Automated short-answer grading (ASAG) remains a challenging task due to the linguistic variability of student responses and the need for nuanced, rubric-aligned partial credit. While Large Language Models (LLMs) offer a promising solution,…

Computation and Language · Computer Science 2026-01-15 Haotian Deng , Chris Farber , Jiyoon Lee , David Tang

As LLM-based systems increasingly operate as agents embedded within human social and technical systems, alignment can no longer be treated as a property of an isolated model, but must be understood in relation to the environments in which…

While the wide adoption of refusal training in large language models (LLMs) has showcased improvements in model safety, recent works have highlighted shortcomings due to the shallow nature of these alignment methods. To this end, the work…

Machine Learning · Computer Science 2026-04-17 Pankayaraj Pathmanathan , Furong Huang

Reasoning methods that adaptively allocate test-time compute have advanced LLM performance on easy to verify domains such as math and code. In this work, we study how to utilize this approach to train models that exhibit a degree of…

Machine Learning · Computer Science 2025-10-28 Taeyoun Kim , Fahim Tajwar , Aditi Raghunathan , Aviral Kumar

Recent advances in AI agents capable of solving complex, everyday tasks, from scheduling to customer service, have enabled deployment in real-world settings, but their possibilities for unsafe behavior demands rigorous evaluation. While…

Artificial Intelligence · Computer Science 2026-02-18 Sanidhya Vijayvargiya , Aditya Bharat Soni , Xuhui Zhou , Zora Zhiruo Wang , Nouha Dziri , Graham Neubig , Maarten Sap