6 papers
Code Monitor Red Teaming for Public-Test-Passing Code
Junchi Liao, Jiawen Deng, Fuji Ren
Visible tests are a common gate for LLM-generated code, but passing them does not certify specification correctness. We study a deployment-like monitoring problem: after code has p…
Auditing Evidence Use in Medical LLM Diagnosis
Junchi Liao, Jiawen Deng, Fuji Ren
Medical LLMs are often evaluated by whether they select the correct diagnosis, but diagnostic accuracy alone does not show whether the model used the case evidence appropriately. W…
Auditing Provenance Sensitivity in LLM Agent Action Selection
Junchi Liao
LLM agents choose tools and arguments from context that mixes user requests, tool outputs, retrieved records, memory, and untrusted text. Evidence can be relevant without being aut…
Evidence Interfaces Shape How Retrieval-Augmented Readers Use Support
Junchi Liao, Jiawen Deng, Fuji Ren
In multi-hop RAG evaluation, a top-k answer score can hide two different failures: the retrieval window may drop part of the support chain, or it may contain support in a form the…
VISTA: Auditing Semantic Divergence in Vision-Language Models
Junchi Liao, Jiawen Deng, Fuji Ren
Vision-language models can exhibit visual concept-conditioned divergence: given images containing demographic features, corporate logos, or ideological symbols, some models produce…
Backdooring CLIP through Concept Confusion
Lijie Hu, Junchi Liao, Weimin Lyu +5
Backdoor attacks pose a serious threat to deep learning models by allowing adversaries to implant hidden behaviors that remain dormant on clean inputs but are maliciously triggered…