25 papers
RLPF: Reinforcement Learning from Performance Feedback for Code Generation
Huihao Jing, Haozhe Cui, Wenbin Hu +9
The paper introduces RLPF, a reinforcement‑learning approach that uses staged performance feedback to train code‑generation models to produce not only correct programs but also fas…
Isolation as a First-Class Principle for LLM-Agent System Safety: Concepts, Taxonomy, Challenges and Future Directions
Huihao Jing, Wenbin Hu, Shaojin Chen +10
The paper surveys how isolating components such as user inputs, tools, execution, inter‑agent communication, and environment can improve safety of LLM‑agent systems, presenting a b…
SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents
Qiao Xiao, Haochen Shi, Yisen Gao +9
Large language model (LLM) agents increasingly rely on agent harnesses that manage context, tools, and multi-turn execution, making tools a central interface for acting in realisti…
Multi-peak structure of meson spectral function in magnetic field
Haoran Li, Ziyue Wang
We investigate the spectral functions of neutral and charged mesons in a hot dense medium under a external magnetic field using the two-flavor quark-meson model within the function…
Into the Gray Zone: Domain Contexts Can Blur LLM Safety Boundaries
Ki Sen Hung, Xi Yang, Chang Liu +7
A central goal of LLM alignment is to balance helpfulness with harmlessness, yet these objectives conflict when the same knowledge serves both legitimate and malicious purposes. Th…
OmniCompliance-100K: A Multi-Domain, Rule-Grounded, Real-World Safety Compliance Dataset
Wenbin Hu, Huihao Jing, Haochen Shi +3
Ensuring the safety and compliance of large language models (LLMs) is of paramount importance. However, existing LLM safety datasets often rely on ad-hoc taxonomies for data genera…