I am an Assistant Professor in Institute of Trustworthy Embodied AI at Fudan University（复旦大学可信具身智能研究院）. I received my bachelor and Ph.D. degree from Department of Computer Science and Technology in Tsinghua University, Beijing, China.

My major research areas are about Embodied AI, multimodel computing, and Information Retrieval. My goal is to build an embodied intelligent system that can perform general human behaviors and interact with humans more effectively. Recently, my interested topics include:

Memory Mechanism of Embodied AI: Developing models that can remember, reason, and use past experiences to interact with real-world physical environments.
Fine-Grained Embodied Manipulation: Realizing robotic atomic manipulation capabilities via reinforcement learning to complete precise, fine-grained physical tasks.
Mulitimodal AI Cognition and Verifiability: Investigating how cognitive abilities and complex behaviors emerge in AI system and whether these capabilities are verifiable to humans.

I am actively looking for self-motivated students to join my research group. If you are passionate about building the future of AI, please feel free to reach out. 本人依托复旦大学可信具身智能研究院和视觉与学习实验室，正积极招募有自驱力的学生加入研究团队，请随时沟通（邮箱：zyye@fudan.edu.cn）。

🔥 News

2026.05, our paper titled How do Human Processes AI-generated Hallucination Contents: a Neuroimaging Study is accepted by ICML 2026.
2026.04, our papers titled Language reconstruction with brain predictive coding from fMRI data and TwinVoice: A Multi-dimensional Benchmark Towards Digital Twins via LLM Persona Simulation are accepted by ACL 2026 main conferences and findings, respectively.
2026.04, our papers titled Individual Turing Test: A Case Study of LLM-based Simulation Using Longitudinal Personal Data and Surge: A benchmark and evaluation framework for scientific survey generation are accepted by SIGIR 2026.
2026.02, our paper titled FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding is accepted by CVPR 2026.
2026.01, our paper titled Robotic Grasping and Placement Controlled by EEG-Based Hybrid Visual and Motor Imagery is accepted by ICRA 2026.
2026.01, our paper titled SLEEP2VEC: Unified Cross-modal Alignment for Heterogeneous Nocturnal Biosignals is accepted by ICLR 2026.
2025.10, I was invited to join the discussion about World Model at CCF YOCSEF. News
2025.10, I gave an invited talk about AI System Empowered by Neuro Feedback at Computational Neuro-Linguistics Forum.Video
2025.08, I gave an invited talk about "Tips for New Researchers" at NLPCC 2025. Slides
2025.08, our paper titled SimVBG: Simulating Individual Values by Backstory Generation is accepted by EMNLP 2025.
2025.05, our paper titled EEG reveals the cognitive impact of polarized content in short video scenarios is available in Nature Scientific Reports.
2025.04, thrilled to announce that I have three co-authered paper accepted by SIGIR 2025, titled Brain Image Reconstruction with Retrieval-Augmented Diffusion, Parametric Retrieval Augmented Generation, and Understanding the Effect of Opinion Polarization in Short Video Browsing.
2025.04, I will attend ICLR 2025 in person for our paper titled Learning LLM-as-a-Judge for Preference Alignment.
2025.03, our paper titled Generative Language Reconstruction from Brain Recordings is published in Nature Communications Biology.
2024.09, our paper titled Pre-trained Model for EEG-based Emotion Recognition won the Best Paper Nomination at CCIR 2024.
2023.12, I gave an invited talk in the Institute of Information in University of Amsterdam about Language Generation from Brain Recordings. Slide
2023.07, I gave an invited talk to Lenovo Inc about Context-based Brain Decoding. Slide
2023.09, I gave an invited talk in Machine Learning Section of Copenhagen University about Brain-Computer Interface for Search. Slide

📝 Publications

Arxiv 2026: VLA-Pro: Cross-Task Procedural Memory Transfer for Vision-Language-Action Models; Shengyu Si, Yuanzhuo Lu, Ruimeng Yang, Ziyi Ye, Zuxuan Wu, Yu-gang Jiang; We develop a cross-task procedural memory transfer framework that yields up to 51\% improvement on existing VLAs.

CVPR 2026: FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding; Yiweng Xie, Bo He, Junke Wang, Xiangyu Zheng, Ziyi Ye, Zuxuan Wu; We propose a novel memory module that can adaptively learn and store streaming video features in a hierarchical manner.

ICRA 2026: Robotic Grasping and Placement Controlled by EEG-Based Hybrid Visual and Motor Imagery; Yichang Liu, Tianyu Wang, Ziyi Ye, Yawei Li, Yanwei Fu, Yu-gang Jiang, Shouyan Wang; We develop a robotic grasping system that can control the placement of objects based on human's mental imagery.

EMNLP 2025: SimVBG: Simulating Individual Values by Backstory Generation; Bangde Du, Ziyi Ye, Zhijing Wu, Monika Jankowska, Shuqi Zhu, Qingyao Ai, Yujia Zhou, Yiqun Liu; We propose a method to simulate individual values by generating backstories as a form of personalization.

SIGIR 2025: Brain Image Reconstruction with Retrieval-Augmented Diffusion; Shuqi Zhu, Ziyi Ye, Yi Zhong, Qingyao Ai, Yujia Zhou, Yiqun Liu; We propose a retrieval augmented diffusion method to reconstruct images with brain input.

ICLR 2025: Learning LLM-as-a-Judge for Preference Alignment; Ziyi Ye, Xiangsheng Li, Qiuchi Li, Qingyao Ai, Yujia Zhou, Wei Shen, Dong Yan, Yiqun Liu; We propose a method to learn a robust and interpretable LLM-as-a-judge for preference alignment.

Nature Communications Biology: Generative Language Reconstruction from Brain Recordings; Ziyi Ye, Qingyao Ai, Yiqun Liu, Maarten de Rijke, Min Zhang, Christina Lioma, and Tuukka Ruotsalo; We propose a method to reconstruct language from brain recordings in a unified generative manner.

See my full publication list.

📖 Experience

Year	Experience
08.2025-	Assistant Professor, Institute of Trustworthy Embodied AI, Fudan University, China.
08.2020-06.2025	Ph.D. student, Department of Computer Science and Technology, Tsinghua University, China.
11.2023-02.2024	Guest Ph.D. student, Institute of Informatics, University of Amsterdam, Netherlands.
07.2023-11.2023	Guest Ph.D. student, Department of Computer Science (DIKU) and the Pioneer Centre for AI, University of Copenhagen, Denmark.
08.2016-07.2020	B.S. student, Department of Computer Science and Technology, Tsinghua University, China.

📝 Academic Service

Journal Reviewer: Nature, Nature Comm. Bio., Nature Dis. Com., ACM TOIS, ACM TKDD, ACM TOMM, SCIS
Conference Senior PC / PC Member: ICLR, NeurIPS, ICML, SIGIR, WWW, ACM MM, CVPR, CIKM, KDD, ACL, EMNLP

-Other Service: CCIR Student Contact (2024.8–2025.8)

🎖 Honors and Awards

2025, Ph.D. Dissertation Award of ACM Shanghai.
2025, Ph.D. Dissertation Award of CIPS.
2025, Outstanding Graduate of Beijing.
2025, Qihang Gold Award, Tsinghua University.
2022, China National Scholarship, Chinese Government.
2017, 2018, 2023, 2024. Overall Excellence Scholarship, Tsinghua University.
2019, Academic Excellence Scholarship, Tsinghua University.
2016, Silver medal in Chinese Mathematical Olympiad(CMO), China Mathematical Federation.

📚 Teaching Experience

2026. Instructor at “Image Processing and Computer Vision”, Fudan University.
2024. Teaching assistant at GenAI summer school, Tsinghua University. (Instructor at “Retrieval-argumented generation”.)
2020-2025. Teaching assistant at “Fundamental Search Engine”, Tsinghua University.

🏆 Competitions

We won the 1st place in TASK 2 (Relevant Statue Retrieval Task) of AILA 2019 (Artificial Intelligence for Legal Assisstance). Working Notes.
We won the 1st place in the NTCIR-15 Micro-activity Retrieval Task. Working Notes.

📊 Datasets

EEG-Hallucination Dataset. In our Paper How do Human Processes AI-generated Hallucination Contents: a Neuroimaging Study.
EEG-Imagenet Dataset. In our Paper EEG-imagenet: An electroencephalogram dataset and benchmarks with image visual stimuli of multi-granularity labels.
EEG-SVC Dataset. In our SIGIR’25 Full paper EEG-svrec: An eeg dataset with user multidimensional affective engagement labels in short video recommendation.
Search-Brainwave Dataset. In our SIGIR’22 Full Paper Why Don’t You Click: Understanding Non-Click Results in Web Search with Brain Signals.
UERCM Dataset. In our The Web Conf’22 Full Paper Towards a Better Understanding of Human Reading Comprehension with Brain Signals.
TianGong-CRL Dataset. In our ASIST’20 Poster Investigating COVID-19-Related Query Logs of Chinese Search Engine Users.

叶子逸（Ziyi Ye）