Publications

2026

1. Preprint

   

   CLSGen: A Dual-Head Fine-Tuning Framework for Joint Probabilistic Classification and Verbalized Explanation

   WonJin Yoon^*, Kangyu Zhu^*, Ian Bulovic, and 5 more authors

   Preprint, 2026
2. AAAI’26 Workshop Best Paper Award

   

   Is Decentralized LLM Agent RL Robust to Heterogeneity? An Asymmetric Tale

   Canyu Chen^*, Kangyu Zhu^*, Zhaorun Chen, and 6 more authors

   In TrustAgent@AAAI 2026, Oral · PerFM@AAAI 2026, Oral, 2026

2025

1. ICML 2025

   

   MMedPO: Aligning Medical Vision-Language Models with Clinical-Aware Multimodal Preference Optimization

   Kangyu Zhu^*, Peng Xia^*, Yun Li, and 3 more authors

   In Proceeding of the Forty-second International Conference on Machine Learning (ICML), 2025
2. NAACL 2025

   

   Guiding Medical Vision-Language Models with Explicit Visual Prompts: Framework Design and Comprehensive Exploration of Prompt Variations

   Kangyu Zhu^*, Ziyuan Qin^*, Huahui Yi, and 4 more authors

   In Proceedings of Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), 2025
3. ICLR 2025

   

   MMed-RAG: Versatile Multimodal RAG System for Medical Vision Language Models

   Peng Xia, Kangyu Zhu, Haoran Li, and 6 more authors

   In Proceedings of International Conference on Learning Representations (ICLR), 2025

   Covered by Banff International Research Station, MarkTechPost, Moonlight Press, neptune.ai.

2024

1. EMNLP 2024

   

   RULE: Reliable Multimodal RAG for Factuality in Medical Vision Language Models

   Peng Xia^*, Kangyu Zhu^*, Haoran Li, and 5 more authors

   In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 2024
2. NeurIPS 2024

   

   CARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language Models

   Peng Xia, Ze Chen, Juanxi Tian, and 8 more authors

   In Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), 2024