Ke Zou

Reliable Federated Disentangling Network for Non-IID Domain Feature

2023 · arXiv (Cornell University)

Federated learning (FL), as an effective decentralized distributed learning approach, enables multiple institutions to jointly train a model without sharing their local data. However, the domain feature shift caused by different acquisition devices/clients substantially degrades …

UniVRSE: Unified Vision-conditioned Response Semantic Entropy for Hallucination Detection in Medical Vision-Language Models

2025 · arXiv (Cornell University)

Vision-language models (VLMs) have great potential for medical image understanding, particularly in Visual Report Generation (VRG) and Visual Question Answering (VQA), but they may generate hallucinated responses that contradict visual evidence, limiting clinical deployment. Although …