I work at the intersection of computer vision and multimodal learning — teaching machines to read the structure and meaning hidden inside images.

Dr. Tao He is a Deputy Researcher at the Laboratory of Intelligent Collaborative Computing, University of Electronic Science and Technology of China (UESTC). Prior to joining UESTC, he was a Research Fellow at Monash University, Australia.

He received his Ph.D. in Computer Science from Monash University in 2022, under the supervision of Associate Professor Yuan-Fang Li, with co-supervision by Professor Lianli Gao. He obtained his M.S. in Computer Science from UESTC in 2018 under Professor Jingkuan Song, and his B.S. in Computer Science from Northeast Normal University (NENU) in 2015.

His research lies in Visual Relation Detection, Multimodal Reasoning, and Embodied AI.

Scene Graph Generation Human–Object Interaction Multimodal Learning Image Retrieval Visual Emotion Recognition Dataset Distillation
Actively recruiting highly self-motivated Ph.D. students — reach out by email.
02

Recent News

NeurIPS
2026
2papers accepted on LLM Reasoning & Incomplete Multimodal Reasoning.
ACMMM
2026
4papers accepted on Visual Relation Detection & Multimodal Reasoning.
ICML
2026
3papers accepted on Dataset Distillation & Large Multimodal Models.
IEEE TMM
1paper accepted on Human–Object Interaction Detection.
CVPR
2026
3papers accepted on Multimodal Learning.
WWW
2026
1paper accepted on Incomplete Multimodal Learning.
AAAI
2026
1paper accepted on Visual Emotion Recognition.
ICCV
2025
2papers accepted on Panoptic SGG & Incomplete Multimodal Learning.
03

Selected Publications

BRACENeurIPS 2026 · Figure
BRACE method framework overview
2026NeurIPS
BRACE: Bipolar Reference-Aware Calibration and Estimation for Incomplete Multimodal Learning
Ruiting Dai, Zesen Cai, XiaoYu Zhao, Yiting Huang, Lisi Mo, Ming Li, Tao He★
PaperCode
C2RICML 2026 · Figure
C2R method framework overview
2026ICML
Mind Your Margin and Boundary: Are Your Distilled Datasets Truly Robust?
Muquan Li, Yingyi Ma, Yihong Huang, Hang Gou, Ke Qin, Ming Li, Yuan-Fang Li, Tao He★
PaperOral~0.7%Code
FlowSGCVPR 2026 · Figure
FlowSG method framework overview
2026CVPR
Can We Build Scene Graphs, Not Classify Them? FlowSG: Progressive Image-Conditioned Scene Graph Generation with Flow Matching
Xin Hu, Ke Qin, Wen Yin, Yuan-Fang Li, Ming Li, Tao He★
Paper
Dataset DistillationCVPR 2026 · Figure
Dataset Distillation method framework overview
2026CVPR
Fixed Anchors Are Not Enough: Dynamic Retrieval and Persistent Homology for Dataset Distillation
Muquan Li, Hang Gou, Yingyi Ma, Rongzheng Wang, Ke Qin, Tao He★
Paper
Anchor DriftWWW 2026 · Figure
Anchor Drift method framework overview
2026WWW
Anchor Drift No More: Hierarchical Consistency-Guided Prompt Distillation for Incomplete Multimodal Learning
Ruiting Dai, Zesen Cai, Lisi Mo, Guiduo Duan, Keren Shi, Tao He★
Paper
TiCALAAAI 2026 · Figure
TiCAL method framework overview
2026AAAI
TiCAL: Typicality-Based Consistency-Aware Learning for Multimodal Emotion Recognition
Wen Yin, Shuai Zhan, Cheng Liu, Xin Hu, Guiduo Duan, Xianzhang Xie, Yuan-Fang Li, Tao He★
PaperCode
SPADEICCV 2025 · Figure
SPADE method framework overview
2025ICCV
SPADE: Spatial-Aware Denoising Network for Open-vocabulary Panoptic Scene Graph Generation with Long- and Local-range Context Reasoning
Xin Hu, Ke Qin, Guiduo Duan, Ming Li, Yuan-Fang Li, Tao He★
PaperCode
Missing-ModalityICCV 2025 · Figure
Missing-Modality method framework overview
2025ICCV
Unbiased Missing-modality Multimodal Learning
Ruiting Dai, Chenxi Li, Yandong Yan, Lisi Mo, Ke Qin, Tao He★
PaperCode
UCDVERCVPR 2025 · Figure
UCDVER method framework overview
2025CVPR
Knowledge-Aligned Counterfactual-Enhancement Diffusion Perception for Unsupervised Cross-Domain Visual Emotion Recognition
Wen Yin, Yong Wang, Guiduo Duan, Dongyang Zhang, Xin Hu, Yuan-Fang Li, Tao He★
PaperCode
Unified SGG–HOIIEEE TIP 2023 · Figure
Unified SGG–HOI method framework overview
2023IEEE TIP
Towards a Unified Transformer-based Framework for Scene Graph Generation and Human-object Interaction Detection
Tao He, Lianli Gao, Jingkuan Song, Yuan-Fang Li
Paper
State-Aware CLIEEE TIP 2022 · Figure
State-Aware CL method framework overview
2022IEEE TIP
State-aware Compositional Learning towards Unbiased Training for Scene Graph Generation
Tao He, Lianli Gao, Jingkuan Song, Yuan-Fang Li
Paper
Open-Vocab SGGECCV 2022 · Figure
Open-Vocab SGG method framework overview
2022ECCV
Towards Open-vocabulary Scene Graph Generation with Prompt-based Finetuning
Tao He, Lianli Gao, Jingkuan Song, Yuan-Fang Li
PaperCode