14 papers
ShadowFPT: Backdooring Federated Prompt Tuning via Shadow Triggers
Kun Zhai, Teng Li, Yunhao Feng +1
Federated Prompt Tuning (FPT) adapts large vision--language models by freezing the pretrained backbone and optimizing only lightweight prompt parameters across clients. Although th…
When Experience Becomes Instruction: Trajectory Poisoning in Self-Evolving Agent Skill Systems
Jialuo Chen, Lingqi Jiang, Xinhao Deng +7
Self-evolving skill (SES) systems distill agent trajectories into persistent skills, allowing untrusted experience to become trusted instruction. We introduce PoisonedEvolution, a…
SkillTrace: Multi-Trace Provenance Auditing for LLM-Agent Skill Reuse
Jialuo Chen, Minghe Wang, Lingqi Jiang +7
LLM-agent ecosystems are rapidly growing around reusable skills: mixed-modality packages of metadata, natural-language instructions, code, tools, references, and operational workfl…
Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification
Yunhao Feng, Ruixiao Lin, Ming Wen +12
LLM agents increasingly perform autonomous actions through external tools, leading to complex and evolving safety risks. However, existing safety testing targets expert-designed sa…
VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills
Ying Li, Yanju Chen, Hongbo Wen +5
Agentic systems increasingly act through third-party skills, allowing model-generated decisions to affect files, communication channels, and cyber-physical devices. These skills of…
Safety in Self-Evolving LLM Agent Systems: Threats, Amplification, and Case Studies
Ruixiao Lin, Xinhao Deng, Qingming Li +12
Self-evolving LLM agent systems, which autonomously update their model parameters, memory, tools, and architectures, introduce a qualitatively new threat landscape in which adversa…
BraveGuard: From Open-World Threats to Safer Computer-Use Agents
Yunhao Feng, Xiaohu Du, Xinhao Deng +13
Computer-use agents extend language models from text generation to sustained interaction with files, terminals, browsers, and external tools. This shift creates safety risks that a…
Constitutional On-Policy Safe Distillation
Ming Wen, Yuxuan Liu, Kun Yang +8
On-policy self-distillation (OPSD) has emerged as an efficient post-training paradigm by using a teacher conditioned on privileged information to provide dense token-level supervis…
SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems
Yunhao Feng, Yifan Ding, Yingshui Tan +6
Skill-based agent systems tackle complex tasks by composing reusable skills, improving modularity and scalability while introducing a largely unexamined security attack surface. We…
Position: AI Safety Requires Effective Controllability
Yige Li, Yunhao Feng, Jun Sun
AI safety is still largely framed as alignment: training models to follow human preferences, safety policies, and normative constraints. That framing has improved the behavior of m…
ASEval: Automated Trajectory-Level Security Testing for Autonomous Agents
Jianan Ma, Xiaohu Du, Ruixiao Lin +9
As autonomous agents (e.g., OpenClaw) increasingly operate with deep system-level privileges to execute complex tasks, they introduce severe, unmitigated security risks. Existing L…
AgentHazard: A Benchmark for Evaluating Harmful Behavior in Computer-Use Agents
Yunhao Feng, Yifan Ding, Yingshui Tan +6
Computer-use agents extend language models from text generation to persistent action over tools, files, and execution environments. Unlike chat systems, they maintain state across…
BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents
Yunhao Feng, Yige Li, Yutao Wu +6
Large language model (LLM) agents execute tasks through multi-step workflows that combine planning, memory, and tool use. While this design enables autonomy, it also expands the at…
PaperAsk: A Benchmark for Reliability Evaluation of LLMs in Paper Search and Reading
Yutao Wu, Xiao Liu, Yunhao Feng +2
Large Language Models (LLMs) increasingly serve as research assistants, yet their reliability in scholarly tasks remains under-evaluated. In this work, we introduce PaperAsk, a ben…