Publications

You can also find my articles on my Google Scholar profile.

2025

MiLDEdit: Reasoning-Based Multi-Layer Design Document Editing

Published in NeurIPS 2025 Workshop on Multimodal Algorithmic Reasoning (Spotlight), 2025

A reasoning-based approach for multi-layer design document editing.

Recommended citation: Zihao Lin, Wanrong Zhu, Jiuxiang Gu, Jihyung Kil, Christopher Tensmeyer, Lin Zhang, Shilong Liu, Lifu Huang, Vlad I Morariu, Tong Sun. "MiLDEdit: Reasoning-Based Multi-Layer Design Document Editing." NeurIPS 2025 Workshop on Multimodal Algorithmic Reasoning, Spotlight paper. (CVPR 2026 under review)
Download Paper

Localizing Knowledge in Diffusion Transformers

Published in NeurIPS 2025, 2025

Localizing knowledge in Diffusion Transformers.

Recommended citation: Arman Zarei, Samyadeep Basu, Keivan Rezaei, Zihao Lin, Sayan Nag, Soheil Feizi. "Localizing Knowledge in Diffusion Transformers." NeurIPS 2025.
Download Paper

Knowing When to Stop: Dynamic Context Cutoff for Large Language Models

Published in NeurIPS 2025, 2025

Dynamic context cutoff for efficient LLM inference.

Recommended citation: Roy Xie, Junlin Wang, Paul Rosu, Chunyuan Deng, Bolun Sun, Zihao Lin, Bhuwan Dhingra. "Knowing When to Stop: Dynamic Context Cutoff for Large Language Models." NeurIPS 2025.
Download Paper

R2I-Bench: Benchmarking Reasoning-Driven Text-to-Image Generation

Published in EMNLP 2025 (Outstanding Paper Award), 2025

A benchmark for evaluating reasoning-driven text-to-image generation.

Recommended citation: Kaijie Chen, Zihao Lin, Zhiyang Xu, Ying Shen, Yuguang Yao, Joy Rimchala, Jiaxin Zhang, Lifu Huang. "R2I-Bench: Benchmarking Reasoning-Driven Text-to-Image Generation." EMNLP 2025, Outstanding Paper Award.
Download Paper

RoRA-VLM: Robust Retrieval-Augmented Vision Language Models

Published in ICCV 2025 Workshop on Knowledge-Intensive Multimodal Reasoning, 2025

Robust retrieval-augmented vision language models.

Recommended citation: Jingyuan Qi, Zhiyang Xu, Rulin Shao, Zihao Lin, Yang Chen, Di Jin, Yu Cheng, Qifan Wang, Lifu Huang. "RoRA-VLM: Robust Retrieval-Augmented Vision Language Models." ICCV 2025 Workshop on Knowledge-Intensive Multimodal Reasoning.
Download Paper

Large Language Models for Conversational User Simulation: A Comprehensive Survey

Published in ARR 2025 July (Under Review), 2025

A comprehensive survey on LLMs for conversational user simulation.

Recommended citation: Bo Ni, Leyao Wang, Yu Wang, Branislav Kveton, et al., Zihao Lin, Trung Bui, Yue Zhao, Tyler Derr, Ryan A Rossi. "Large Language Models for Conversational User Simulation: A Comprehensive Survey." ARR 2025 July, Under Review.
Download Paper

Persona-SQ: An end-to-end Framework for Personalized Suggested Questions Generation for Long Documents

Published in NAACL 2025 Demo Track, 2025

An end-to-end framework for generating personalized suggested questions for long documents.

Recommended citation: Zihao Lin, Zichao Wang, Yuanting Pan, Varun Manjunatha, Ryan Rossi, Angela Lau, Lifu Huang, Tong Sun. "Persona-SQ: An end-to-end Framework for Personalized Suggested Questions Generation for Long Documents." NAACL 2025, Demo Track.
Download Paper

A Survey on Explainability for the Inner Mechanism of Multi-Modal Foundation Models

Published in arXiv, 2025

A comprehensive survey on explainability for multi-modal foundation models.

Recommended citation: Zihao Lin, Samyadeep Basu, Mohammad Beigi, et al., Lifu Huang. "A Survey on Explainability for the Inner Mechanism of Multi-Modal Foundation Models." arXiv:2502.17516.
Download Paper

2024

ToolBeHonest: A Multi-level Hallucination Diagnostic Benchmark for Tool-Augmented LLMs

Published in EMNLP 2024, 2024

A multi-level hallucination diagnostic benchmark for tool-augmented LLMs.

Recommended citation: Yuxiang Zhang, Jing Chen, Junjie Wang, Yaxin Liu, Cheng Yang, Zihao Lin, et al., Hayato Yamana. "ToolBeHonest: A Multi-level Hallucination Diagnostic Benchmark for Tool-Augmented LLMs." EMNLP 2024.
Download Paper

MMedAgent: Learning to Use Medical Tools with Multi-modal Agent

Published in EMNLP 2024 Findings, 2024

A multi-modal medical agent that learns to use medical tools.

Recommended citation: Binxu Li, Tiankai Yan, Yuanting Pan, Zhe Xu, Jie Luo, Ruiyang Ji, Shilong Liu, Haoyu Dong*, Zihao Lin*, Yixin Wang* (* co-corresponding authors). "MMedAgent: Learning to Use Medical Tools with Multi-modal Agent." EMNLP 2024 Findings.
Download Paper

InternalInspector: Robust Confidence Estimation in LLMs through Internal States

Published in EMNLP 2024 Findings, 2024

Robust confidence estimation in LLMs through internal states.

Recommended citation: Mohammad Beigi, Ying Shen, Runing Yang, Zihao Lin, et al., Lifu Huang. "InternalInspector: Robust Confidence Estimation in LLMs through Internal States." EMNLP 2024 Findings.
Download Paper

Holistic Evaluation for Interleaved Text-and-Image Generation

Published in EMNLP 2024, 2024

Holistic evaluation for interleaved text-and-image generation.

Recommended citation: Minqian Liu, Zhiyang Xu, Zihao Lin, Trevor Ashby, Joy Rimchala, Jiaxin Zhang, Lifu Huang. "Holistic Evaluation for Interleaved Text-and-Image Generation." EMNLP 2024.
Download Paper

Rethinking the Uncertainty: A Critical Review and Analysis in the Era of Large Language Models

Published in arXiv, 2024

A critical review of uncertainty quantification in LLMs.

Recommended citation: Mohammad Beigi, Sijia Wang, Ying Shen, Zihao Lin, et al., Lifu Huang. "Rethinking the Uncertainty: A Critical Review and Analysis in the Era of Large Language Models." arXiv:2410.20199.
Download Paper

Navigating the Dual Facets: A Comprehensive Evaluation of Sequential Memory Editing in Large Language Models

Published in ACL 2024, 2024

A comprehensive evaluation of sequential memory editing methods in LLMs.

Recommended citation: Zihao Lin, Hongxuan Li, Yufan Zhou, Yuxiang Zhang, Mohammad Beigi, Lifu Huang. "Navigating the Dual Facets: A Comprehensive Evaluation of Sequential Memory Editing in Large Language Models." Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024).
Download Paper

2023

Efficient Federated Prompt Tuning for Black-Box Large Pre-trained Models

Published in arXiv, 2023

An efficient approach for federated prompt tuning of black-box large pre-trained models.

Recommended citation: Zihao Lin*, Yan Sun*, Yifan Shi, Xueqian Wang, Lifu Huang, Li Shen, Dacheng Tao (*equal contribution). "Efficient Federated Prompt Tuning for Black-Box Large Pre-trained Models." arXiv:2310.03123.
Download Paper

Rethinking Medical Report Generation: Disease Revealing Enhancement with Knowledge Graph

Published in ICML 2023, 2023

A knowledge graph enhanced approach for medical report generation.

Recommended citation: Zihao Lin*, Yixin Wang*, Haoyu Dong* (* equal contribution). "Rethinking Medical Report Generation: Disease Revealing Enhancement with Knowledge Graph." Proceedings of the 40th International Conference on Machine Learning (ICML 2023).
Download Paper

Towards More Suitable Personalization in FL via Decentralized Partial Model Training

Published in arXiv, 2023

Personalization in federated learning via decentralized partial model training.

Recommended citation: Yifan Shi, Yingqi Liu, Yan Sun, Zihao Lin, Li Shen, Xueqian Wang, Dacheng Tao. "Towards More Suitable Personalization in FL via Decentralized Partial Model Training." arXiv:2305.15157.
Download Paper

Prompt Tuning based Adapter for Vision-Language Model Adaption

Published in arXiv, 2023

Prompt tuning based adapter for vision-language model adaptation.

Recommended citation: Jingchen Sun, Jiayu Qin, Zihao Lin, Changyou Chen. "Prompt Tuning based Adapter for Vision-Language Model Adaption." arXiv:2303.15234.
Download Paper

Trust It or Not: Confidence-Guided Automatic Radiology Report Generation

Published in Neurocomputing, 2023

A confidence-guided approach for automatic radiology report generation.

Recommended citation: Zihao Lin*, Yixin Wang*, Zhe Xu, Haoyu Dong, Jiang Tian, Jie Luo, Zhongchao Shi, Lifu Huang, Yang Zhang, Jianping Fan, and Zhiqiang He (* equal contribution). "Trust It or Not: Confidence-Guided Automatic Radiology Report Generation." Neurocomputing, 127374.
Download Paper

2021

ACN: Adversarial Co-training Network for Brain Tumor Segmentation with Missing Modalities

Published in MICCAI 2021, 2021

Adversarial co-training network for brain tumor segmentation with missing modalities.

Recommended citation: Wang Yixin, Yang Zhang, Yang Liu, Zihao Lin, et al., Zhiqiang He. "ACN: Adversarial Co-training Network for Brain Tumor Segmentation with Missing Modalities." MICCAI 2021.
Download Paper