5 papers
HARDE: Optimizing Agent Harnesses for Runtime Risk Detection and Execution Control
Zhuo Liu, Moxin Li, Zhixin Ma +3
Large language model (LLM) agents are vulnerable to safety risks such as injected malicious instructions or misleading information, motivating runtime defenses that prevent unsafe…
RISE-RL: Rubric-Informed Selective Exploration for Open-Ended Reinforcement Learning
Jinkun Hou, Zhuo Liu, Huimin Ren +3
Aligning Large Language Models (LLMs) for open-ended tasks is challenging because responses must satisfy multidimensional criteria without following a single correct generation tra…
TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards
Xiqiao Xiong, Ouxiang Li, Zhuo Liu +5
Large language models have seen widespread adoption, yet they remain vulnerable to multi-turn jailbreak attacks, threatening their safe deployment. This has led to the task of trai…
Assistant-Guided Mitigation of Teacher Preference Bias in LLM-as-a-Judge
Zhuo Liu, Moxin Li, Xun Deng +2
LLM-as-a-Judge employs large language models (LLMs), such as GPT-4, to evaluate the quality of LLM-generated responses, gaining popularity for its cost-effectiveness and strong ali…
Self-Improvement Towards Pareto Optimality: Mitigating Preference Conflicts in Multi-Objective Alignment
Moxin Li, Yuantao Zhang, Wenjie Wang +4
Multi-Objective Alignment (MOA) aims to align LLMs' responses with multiple human preference objectives, with Direct Preference Optimization (DPO) emerging as a prominent approach.…