← Search

Zhiwei Jin

2 accepted papers

2026

Learning from Fine-Grained Visual Discrepancies: Mitigating Multimodal Hallucinations via In-Context Visual Contrastive Optimization

ICML 2026poster

Multimodal hallucination remains a persistent challenge for Vision-Language Models (VLMs). Standard textual Direct Preference Optimization (DPO) often fails to mitigate it due to a lack of explicit visual supervision. While existing works introduce visual preference DPO by contrasting original image…

Cited by 0SourceScholar
2023

Learn over Past, Evolve for Future: Forecasting Temporal Trends for Fake News Detection

ACL 2023industry

Fake news detection has been a critical task for maintaining the health of the online news ecosystem. However, very few existing works consider the temporal shift issue caused by the rapidly-evolving nature of news data in practice, resulting in significant performance degradation when training on p…