5 papers
When Code Becomes an LLM Call: An Empirical Study of Call-Site Contracts in Agentic Systems
Shihao Weng, Longtian Wang, Yang Feng +6
Software quality has long rested on specifications that state what code must do, and on testing and verification that check that the code does it. Agentic systems change this found…
AgentAntibody: An Adaptive Immune System for Defending LLM Agents against Prompt Injection
Shihao Weng, Yang Feng, Xiaofei Xie +1
Prompt injection remains a critical threat to LLM agents, yet existing defenses treat each task as a self-contained problem, independent of previous encounters. In practice, user r…
Beyond Accuracy: Policy Invariance as a Reliability Test for LLM Safety Judges
Shihao Weng, Yang Feng, Xiaofei Xie
LLM-as-a-Judge pipelines have become the de facto evaluator for agent safety, yet existing benchmarks treat their verdicts as ground-truth proxies without checking whether the verd…
ARGUS: Defending LLM Agents Against Context-Aware Prompt Injection
Shihao Weng, Yang Feng, Jinrui Zhang +3
Large Language Model (LLM) agents are increasingly deployed as task-oriented software systems that use runtime context to decide and act on behalf of users. This delegation model m…
AtPatch: Debugging Transformers via Hot-Fixing Over-Attention
Shihao Weng, Yang Feng, Jincheng Li +3
Transformer-based deep neural networks (DNNs) affected by backdoor attacks and unfairness typically exhibit anomalous attention patterns, leading to over-attend to backdoor trigger…