Ailing Zeng 曾爱玲

Biography

I am the Head of Foundation Video Model at Bilibili, building the frontier audio-video generation foundation models.

Video is becoming the universal language of expression and information — for humans to understand and enjoy, and for machines to see and learn about the world.

Foundation models will turn video from a medium we record into a medium we create: new forms of AI-native entertainment for people, and world simulators for agents and robots.

We are only at the beginning — the upper bound of multimodal content intelligence and quality remains wide open.

Previously, I led teams on interactive video generation and human-centric research at Anuttacon, Tencent, and IDEA, after my Ph.D. at The Chinese University of Hong Kong (advised by Prof. Qiang Xu) and a visiting stay at the Robotics Institute, CMU. Several works from my teams — LPM, DWPose, Grounded-SAM, Motion-X — are widely adopted by the community; see Google Scholar for the full list.

We are hiring! My team at Bilibili is looking for both technical and creative content talents. We are a flat, high-agency organization, pursuing the upper bound of multimodal content intelligence and quality. Feel free to reach out via email.

News

[2026.04] We introduce LPM 1.0, a video-based character performance model that generates real-time video with full-duplex conversation, identity-consistent infinite-length generation, and nuanced human-like performance: [Project Page] [Paper].
[2026.01] We are hosting the ICLR 2026 Tutorial on AI with Recursive Self-Improvement: [Link].
[2025] 5 papers were accepted to ICCV/CVPR/AAAI/ICML 2025, and 4 papers were accepted to TPAMI.
[2025.10] Invited talk at the 1st Workshop on Interactive Human-centric Foundation Models, ICCV 2025.
[2025.02] Invited talk on video generation at the Max Planck Institute for Intelligent Systems, hosted by Michael J. Black.
[2024.10] I serve as an Area Chair for CVPR 2025.
[2024.05] LTSF-Linear was selected as the most Influential Paper in AAAI 2023.
[2023 - 2024] 26 papers were accepted to CVPR/ICML/ICLR/ECCV/ICCV/AAAI/SIGGRAPH-Asia/NeurIPS.

Selected Work

A few representative works; see the full list at Google Scholar. (*equal contribution, ^#corresponding author or project lead)

LPM 1.0: Video-based Character Performance Model [Project Page]
Ailing Zeng^#, Casper Yang, Chauncey Ge, Eddie Zhang, Garvey Xu, Gavin Lin, Gilbert Gu, Jeremy Pi, Leo Li, Mingyi Shi, Sheng Bi, Steven Tang, Thorn Hang, Tobey Guo, Vincent Li, Xin Tong^#, Yikang Li, Yuchen Sun, Yue (R) Zhao, Yuhan Lu, Yuwei Li, Zane Zhang, Zeshi Yang, Zi Ye
technical report, 2026

Real-time video generation with full-duplex conversation, identity-consistent infinite-length generation, and nuanced human-like performance.

The Dawn of Video Generation: Preliminary Explorations with SORA-like Models [Website]
Ailing Zeng^*, Yuhang Yang^*, Weidong Chen, Wei Liu
technical report, 2024

A systematic empirical study on 21 SORA-like text-to-video, image-to-video, and video-to-video models.

Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks [Code]
Tianhe Ren, Shilong Liu, Ailing Zeng, et al.
technical report, 2024

An open-world visual foundation pipeline combining SAM with grounding models, widely used across detection, segmentation, and generation tasks.

DWPose: Effective Whole-body Pose Estimation with Two-stages Distillation [Code]
Zhendong Yang, Ailing Zeng, Chun Yuan, Yu Li
ICCV Workshop, 2023

A better alternative to OpenPose; a standard conditioning module in many controllable video generation pipelines.

Motion-X: A Large-scale 3D Expressive Whole-body Human Motion Dataset [Homepage] [Code]
Jing Lin*, Ailing Zeng*^#, Shunlin Lu*, et al.
NeurIPS Datasets and Benchmarks Track, 2023

A large-scale 3D expressive whole-body human motion dataset (over 10M frames) with SMPL-X, text, audio, and RGB modalities.

Are Transformers Effective for Time Series Forecasting? [Code]
Ailing Zeng, Muxi Chen, Lei Zhang, Qiang Xu
(Oral) AAAI, 2023 — the most Influential Paper in AAAI 2023

Questioning the validity of Transformer-based time series forecasting solutions via a simple linear baseline.

Honors & Awards

KAUST AI Rising Star, 2025
Shenzhen Artificial Intelligence Natural Science Award, 2023
Shenzhen Pengcheng special talent award, 2023
Full Postgraduate Studentship, CUHK (2017 - 2021)
Excellent League Member, Top 1% student of Xiamen University
National Scholarship (2015)