8 papers
Control Under Compression: Reliability Frontiers for Tool-Using Agents
Yinghan Hou, Zongyou Yang
Tool-using language-model agents are governed not only by task prompts but also by persistent system-side instructions that specify tools, arguments, policies, execution protocols,…
VeraRAN: Pre-Actuation Certification and Event-Causal Synchronization Repair for Asynchronous Multi-Interface RAN Plans
Yinghan Hou, Zongyou Yang
Agentic RAN controllers combine mobility, energy, and resource actions across independently implemented interfaces. Even when each command is valid and the target state is safe, as…
Accuracy Hides How Language Models Fail: Measuring Failure States Under Matched Output Budgets
Zongyou Yang, Yinghan Hou
Language-model benchmarks collapse two distinct measurement questions into a single accuracy score: whether a response reached an evaluable state, and whether its answer was judged…
When the Judge Changes, So Does the Measurement: Auditing LLM-as-Judge Reliability
Zongyou Yang, Yinghan Hou, Xiaokun Yang
An LLM-as-judge score can move even when the candidate responses stay fixed, simply because the evaluator has changed. We treat this evaluator-replacement ambiguity as a measuremen…
Degradation-Consistent Paired Training for Robust AI-Generated Image Detection
Zongyou Yang, Yinghan Hou, Xiaokun Yang
AI-generated image detectors suffer significant performance degradation under real-world image corruptions such as JPEG compression, Gaussian blur, and resolution downsampling. We…
SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills
Yinghan Hou, Zongyou Yang, Zaihu Pang +1
Agent skills combine natural-language instructions with executable code while inheriting an agent's filesystem, credential, and network access. Attacks can span prose and files, wh…