Miao Liu

Tsinghua University · College of Artificial Intelligence

Miao Liu 刘淼

Assistant Professor

Previously Research Scientist at Meta GenAI

Jaime Lannister on horseback, charging alone from the left toward Daenerys and her enormous dragon on the right
A personal note on doing science

This image of Jaime Lannister charging alone at Daenerys and her dragon reveals what it often takes to do science—you must be willing to stand as the lonely warrior.

Miao Liu · 刘淼

MEOW Lab — Modeling Egocentric Omni World

About me

I’m an Assistant Professor at Tsinghua University’s College of Artificial Intelligence, where I lead the MEOW (Modeling Egocentric Omni World) Lab.

Designing human-centered AI that sees through your eyes, learns your skills, and understands your intentions.构建能“看你所见、学你所会、懂你所想”的下一代人本智能系统。

I received my Ph.D. in Robotics from Georgia Tech, advised by Prof. James Rehg, and work closely with Prof. Yin Li at the University of Wisconsin–Madison. I was fortunate to collaborate with Prof. Siyu Tang and Prof. Michael Black during visits to ETH Zurich and the Max Planck Institute.

Previously, I was a Research Scientist at Meta GenAI, focusing on egocentric vision and generative AI. During my internship at Facebook Reality Labs, I worked with Dr. Chao Li, Dr. Lingni Ma, Dr. Kiran Somasundaram, and Prof. Kristen Grauman on egocentric action recognition and localization.

I have been honored with Best Paper nominations at CVPR 2022 and ECCV 2024, and the BMVC 2022 Best Student Paper Award. As a primary contributor, I helped build Ego4D, Ego-Exo4D, EGTEA Gaze+, and BEHAVIOR Vision Suite. I also designed models intended for deployment in next-generation smart glasses developed by Meta Reality Labs. At Meta GenAI, I contributed to training and evaluating large-scale generative multimodal models, including Emu and the multimodal components of Llama 3 and Llama 4.

Our research north star

Human-centered intelligence

We connect human-centric perception, multimodal foundation models, embodied intelligence, and cognitive science to build AI that understands people, adapts to individuals, and proactively assists in daily life.

01

Egocentric vision

Understanding people through first-person perception of their actions, attention, and interactions with the world.

02

Embodied AI

Enabling robots to understand, learn from, and assist people in dynamic, human-centered environments.

03

Personalized multimodal models

Building foundation models that remember user history, model individual preferences, and provide personalized assistance.

04

AI with theory of mind

Developing agents that reason about intentions, beliefs, and cognitive states to anticipate human needs.

MEOW Lab research north star: connecting human-centered perception, foundation models, embodied intelligence, and cognitive science
Our research vision · Click to view the full figure

盼诸君助我!
My group is always looking for talented students interested in human-centered AI.

Get in touch ↗

From the lab

Latest news

Since joining Tsinghua · August 2025 onwards. Scroll for earlier updates.

  • Sep. 2026: Two papers accepted to NeurIPS 2026.
  • Aug. 2026: Three papers accepted to EMNLP 2026 (1 main + 1 oral + 1 findings).
  • Jun. 2026: Three papers accepted to ECCV 2026.
  • May 2026: One paper accepted to ICML 2026.
  • Apr. 2026: One paper accepted to ACL 2026 main conference as an oral presentation.
  • Feb. 2026: Two papers accepted to CVPR 2026 (1 main + 1 findings).
  • Jan. 2026: One paper accepted to WACV 2026.
  • Jan. 2026: One paper accepted to ICLR 2026.
  • Sep. 2025: One paper accepted to the NeurIPS 2025 Datasets and Benchmarks Track.
  • Aug. 2025: One paper accepted to EMNLP Findings 2025.
  • Aug. 2025: Officially left Meta and joined Tsinghua University, College of AI.
Earlier news · 2022–2025
  • Jun. 2025: Received the Egocentric Vision (EgoVis) 2023/2024 Distinguished Paper Awards .
  • Feb. 2025: Three papers accepted to CVPR 2025.
  • Oct. 2024: Our LEGO paper has been nominated as one of the 15 award candidates at ECCV 2024.
  • Jul. 2024: Two corresponding-author papers accepted to ECCV 2024 (1 Poster, 1 Oral).
  • Jun. 2024: Received the Egocentric Vision (EgoVis) 2022/2023 Distinguished Paper Awards
  • Feb. 2024: Three papers accepted to CVPR 2024 (1 Poster, 1 Highlight, 1 Oral).
  • Nov. 2023: One paper accepted to IEEE TPAMI.
  • Nov. 2023: One paper accepted to IJCV.
  • Jun. 2023: One paper accepted to ACL 2023 in Findings.
  • Nov. 2022: Our paper on Egocentric Gaze Estimation won the Best Student Paper Prize at BMVC 2022!
  • Sep. 2022: One paper accepted to BMVC 2022 for spotlight presentation!
  • Aug. 2022: I started my new journey at Meta Reality Labs.

Research in print

Selected publications

Full list on Google Scholar ↗
2026 NeurIPS ED

IPIBench: Evaluating Interactive Proactive Intelligence of MLLMs under Continuous Streams

Jinzhao Li, Yinuo Chen, Wenxuan Song, Yijia Lei, Yichi Zhang, Honglei Yan, Panwang Pan, Miao Liu .

Conference on Neural Information Processing Systems (NeurIPS) 2026 · ED Track.

2026 EMNLP

PIVOTSBench: Evaluating Fine-Grained Interpersonal Relationship Reasoning in Multimodal Large Language Models

Shuxiang Zhang, Yiting Yin, Wenxuan Song, Yuhang Wu, Miao Liu .

Conference on Empirical Methods in Natural Language Processing (EMNLP) 2026. ( Oral )

2026 EMNLP

MeetingToM: Evaluating Multimodal LLMs on Theory-of-Mind Reasoning in Multi-Party Meetings

Ziyi Wang*, Yuhang Wu*, Dongxu Piao, Xingyu Liu, Tianhui Zhou, Miao Liu .

Conference on Empirical Methods in Natural Language Processing (EMNLP) 2026. *: Equal Contribution

2026 ECCV

EgoSAT: A Comprehensive Benchmark of Egocentric Streaming Interaction Understanding

Yijia Lei, Jinzhao Li, Yichi Zhang, Jiacheng Hua, Yin Li, Miao Liu .

The European Conference on Computer Vision (ECCV) 2026.

2026 ECCV

PIPBench: A Profile-Inclusive Framework for Personalized Image Generation Evaluation

Yuhang Wu, Shuxiang Zhang, WEE HIAN CHING, Chi Zhang, Miao Liu .

The European Conference on Computer Vision (ECCV) 2026.

2026 ICML

GameVerse: Can Vision-Language Models Learn from Video-based Reflection?

Kuan Zhang, Dongchen Liu, Qiyue Zhao, Jinkun Hou, Xinran Zhang, Qinlei Xie, Miao Liu† , Yiming Li†.

The International Conference on Machine Learning (ICML) 2026. †: Co-corresponding Author

2026 ACL

Unleashing Spatial Reasoning in Multimodal Large Language Models via Textual Representation Guided Reasoning

Jiacheng Hua, Yishu Yin, Yuhang Wu, Tai Wang, Yifei Huang, Miao Liu .

Annual Meeting of the Association for Computational Linguistics (ACL) 2026. ( Oral , acceptance rate 3.3%)

2026 CVPR

EgoProx: Evaluating MLLMs on Egocentric 3D Proximity Reasoning Across a Cognitive Hierarchy

Jinzhao Li, Yinuo Chen*, Dongxu Piao*, Panwang Pan, Yifan Yu, Dong Wang, Honglei Yan, Liang Yue, Shaofei Wang, Yixin Chen, Siyuan Huang, Miao Liu .

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026.

2025 NeurIPS

In the Eye of MLLM: Benchmarking Egocentric Video Intent Understanding with Gaze-Guided Prompting

Taiying Peng, Jiacheng Hua, Miao Liu† , Feng Lu†.

Conference on Neural Information Processing Systems (NeurIPS) 2025. †: Co-corresponding Author

2025 CVPR

Building a Mind Palace: Structuring Environment-Grounded Semantic Graphs for Effective Long Video Analysis with LLMs

Zeyi Huang*, Yuyang Ji*, Xiaofang Wang, Nikhil Mehta, Tong Xiao, Donghyun Lee, Sigmund Vanvalkenburgh, Shengxin Zha, Bolin Lai, Licheng Yu, Ning Zhang, Yong Jae Lee†, Miao Liu† .

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025. †: Co-corresponding Author

2025 CVPR

Unleashing In-context Learning of Autoregressive Models for Few-shot Image Manipulation

Bolin Lai, Felix Juefei-Xu, Miao Liu , Xiaoliang Dai, Nikhil Mehta, Chenguang Zhu, Zeyi Huang, James M. Rehg, Sangmin Lee, Ning Zhang, Tong Xiao.

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025. ( Highlight )

2024 ECCV

LEGO: Learning EGOcentric Action Frame Generation via Visual Instruction Tuning

Bolin Lai, Xiaoliang Dai, Lawrence Chen, Guan Pang, James M. Rehg, Miao Liu .

European Conference on Computer Vision (ECCV) 2024. ( Oral , Best Paper Award Candidate 15/8585 )

2024 ECCV

Listen to Look into the Future: Audio-Visual Egocentric Gaze Anticipation

Bolin Lai, Fiona Ryan, Wenqi Jia, Miao Liu† , James M. Rehg†.

European Conference on Computer Vision (ECCV) 2024. †: Co-corresponding Author

2024 CVPR

BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation

Yunhao Ge*, Yihe Tang*, Jiashu Xu*, Cem Gokmen*, Chengshu Li, Wensi Ai, Benjamin Jose Martinez, Arman Aydin, Mona Anvari, Ayush K Chakravarthy, Hong-Xing Yu, Josiah Wong, Sanjana Srivastava, Sharon Lee, Shengxin Zha, Laurent Itti, Yunzhu Li, Roberto Martín-Martín, Miao Liu , Pengchuan Zhang, Ruohan Zhang, Li Fei-Fei, Jiajun Wu.

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2024 ( Spotlight ) *: Equal Contribution

2024 CVPR

Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives

Kristen Grauman et al. (including Miao Liu).

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2024 ( Oral )

2024 IJCV

In the Eye of Transformer: Global–Local Correlation for Egocentric Gaze Estimation and Beyond

Bolin Lai, Miao Liu† , Fiona Ryan, James M. Rehg.

International Journal of Computer Vision (IJCV), 2024; first published online in 2023. †: Student Mentor

2022 BMVC

In the Eye of Transformer: Global-Local Correlation for Egocentric Gaze Estimation

Bolin Lai, Miao Liu† , Fiona Ryan, James M. Rehg.

British Machine Vision Conference (BMVC) 2022. †: Student Mentor, Co-corresponding Author ( Spotlight , Best Student Paper Prize )

2022 CVPR

Ego4D: Around the World in 3,000 Hours of Egocentric Video

Kristen Grauman et al. (including Miao Liu).

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022 ( Oral , best paper finalist, 33/8161 ) Key driver for Social Benchmark and Forecasting Benchmark

2020 BMVC

Attention Distillation for Learning Video Representations

Miao Liu , Xin Chen, Yun Zhang, Yin Li, and James M. Rehg.

British Machine Vision Conference (BMVC) 2020 ( Oral , acceptance rate 5.0%)

2020 ECCV

Forecasting Human-Object Interaction: Joint Prediction of Motor Attention and Actions in First Person Video

Miao Liu , Siyu Tang, Yin Li, and James M. Rehg.

European Conference on Computer Vision (ECCV) 2020 ( Oral , acceptance rate 2.0%)

The people behind the research

Students

Ph.D. students · MEOW Lab

In the classroom

Teaching

SPRING
2026

Multimodal Generative AI System Design

Course 80940012 · Tsinghua University

Let’s connect

Say hello.

For research collaborations and prospective student inquiries, please reach out by email.

College of Artificial Intelligence
Tsinghua University · Beijing, China