Hanwang Zhang
Professor
Works on computer vision and multimodal machine reasoning, including causal inference for vision-language learning and compositional visual reasoning.
Hanwang Zhang's lab at NTU works on vision-language & multimodal, multimodal & grounded language and reasoning & agents. 85 papers since 2023. 9 PhD students and 1 master's student. 16 alumni.
At a glance
- papers, 2023–2026
- 22/ year
- at T1 venues
- 0%
- PhD students
- 9
- MS students
- 1
- alumni listed
- 16
Papers by yearfrom the lab homepage, 2023–2026
- 16 papers2026 · T1 0 · collecting
- 20 papers2025 · T1 0
- 30 papers2024 · T1 0
- 19 papers2023 · T1 0
T1 venuesOther
Research focuspaper titles that use each topic's words
- Vision-language & multimodal22 papers
- Multimodal & grounded language19 papers
- Reasoning & agents4 papers
- reasoning3 papers
- compositional1 paper
Paper list
From the lab homepage.
202616 papers
- Proceedings of the IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRAdapting Point Cloud Analysis via Multimodal Bayesian Distribution Learning
- Proceedings of the IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRDeepAlign: Mitigating Modality Conflict through Modality-Specific Alignment
- The AAAI Conference on Artificial Intelligence. AAAIDEPO: Dual-Efficiency Preference Optimization for LLM Agents
- The AAAI Conference on Artificial Intelligence. AAAIDragNeXt: Rethinking Drag-Based Image Editing
- The AAAI Conference on Artificial Intelligence. AAAIHierarchical Semantic Alignment for Image Clustering
- The International Conference on Learning Representations. ICLRLook Carefully: Adaptive Visual Reinforcements in Multimodal Large Language Models for Hallucination Mitigation
- Proceedings of the IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRModeling Cross-vision Synergy for Unified Large Vision Model
- The AAAI Conference on Artificial Intelligence. AAAINeuSpring: Neural Spring Fields for Reconstruction and Simulation of Deformable Objects from Videos
- The AAAI Conference on Artificial Intelligence. AAAI OralObject Fusion via Diffusion Time-step for Customized Image Editing with Single Example
- The AAAI Conference on Artificial Intelligence. AAAIPersonalize Your Gaussian: Consistent 3D Scene Personalization from a Single Image
- Proceedings of the IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRPrincipled Steering via Null-space Projection for Jailbreak Defense in Vision-Language Models
- The AAAI Conference on Artificial Intelligence. AAAIPushing Rendering Boundaries: Hard Gaussian Splatting
202520 papers
- International Conference on Machine Learning. ICML3D Question Answering via only 2D Vision-Language Models
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPR OralAnyEdit: Mastering Unified High-Quality Image Editing for Any Idea
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPR HighlightCARE Transformer: Mobile-Friendly Linear Visual Transformer via Decoupled Dual Interaction
- Conference on Neural Information Processing Systems. NeurIPS SpotlightCo-Reinforcement Learning for Unified Multimodal Understanding and Generation
- International Conference on Computer Vision. ICCVCorvid: Improving Multimodal Large Language Models Towards Chain-of-Thought Reasoning
- International Conference on Machine Learning. ICMLDiffusion Model with Causal Generation and Cache Sharing
- International Conference on Computer Vision. ICCVDistilling Parallel Gradients for Fast ODE Solvers of Diffusion Models
- International Conference on Computer Vision. ICCVDynamic Multimodal Prototype Learning in Vision-Language Models
- Conference on Neural Information Processing Systems. NeurIPS SpotlightEnhancing CLIP Robustness via Cross-Modality Alignment
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPR OralGenerative Multimodal Pretraining with Discrete Diffusion Timestep Tokens
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPR HighlightLearning 4D Panoptic Scene Graph Generation from Rich 2D Visual Scene
- International Conference on Computer Vision. ICCVNautilus: Locality-aware Autoencoder for Scalable Mesh Generation
202430 papers
- Conference on Neural Information Processing Systems. NeurIPSAction Imitation in Common Action Space for Customized Action Image Synthesis
- The International Conference on Machine Learning. ICML SpotlightAuto-Encoding Morph-Tokens for Multimodal LLM
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRClasses Are Not Equal: An Empirical Study on Image Recognition Fairness
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRConsistent3D: Towards Consistent High-Fidelity Text-to-3D Generation with Deterministic Sampling Prior
- Conference on Neural Information Processing Systems. NeurIPSDecoupled Kullback-Leibler Divergence Loss
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRDiffusion Time-step Curriculum for One Image to 3D Generation
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRDisCo: Disentangled Control for Realistic Human Dance Generation
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRDiscriminative Probing and Tuning for Text-to-Image Generation
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRDistributionally Generative Augmentation for Fair Facial Attribute Classification
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRDoubly Abductive Counterfactual Inference for Text-based Image Editing
- The AAAI Conference on Artificial Intelligence. AAAIDual-Perspective Knowledge Enrichment for Semi-Supervised 3D Object Detection
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPREmpowering Dynamics-aware Text-to-Video Diffusion with LLMs
202319 papers
- The IEEE/CVF Computer Vision and Pattern Recognition Conference. CVPRBootstrap Your Own Prior: Towards Distribution-Agnostic Novel Class Discovery
- The Eleventh International Conference on Learning Representations. ICLRCompositional Prompt Tuning with Motion Cues for Open-vocabulary Video Relation Detection
- Annual Meeting of the Association for Computational Linguistics. ACLCounterfactual Active Learning for Out-of-Distribution Generalization
- IEEE Transactions on Pattern Analysis and Machine Intelligence. TPAMICounterfactual Samples Synthesizing and Training for Robust Visual Question Answering
- The AAAI Conference on Artificial Intelligence. AAAIDebiased Fine-Tuning for Vision-language Models by Prompt Regularization
- International Conference on Computer Vision. ICCVEquivariant Similarity for Vision-Language Foundation Models
- Conference on Neural Information Processing Systems. NeurIPSGeneralized Logit Adjustment: Calibrating Fine-tuned Models by Removing Label Bias in Foundation Models
- Annual Meeting of the Association for Computational Linguistics. ACL FindingsHypothetical Training for Robust Machine Reading Comprehension of Tabular Context
- Conference on Neural Information Processing Systems. NeurIPSImagine That! Abstract-to-Intricate Text-to-Image Synthesis with Scene Graph Hallucination Diffusion.
- International Conference on Computer Vision. ICCVInvariant Feature Regularization for Fair Face Recognition
- International Conference on Computer Vision. ICCVInvariant Training 2D-3D Joint Hard Samples for Few-Shot Point Cloud Recognition
- International Conference on Computer Vision. ICCVLearning Trajectory-Word Alignments for Video-Language Tasks
Where alumni went16 listed
Apply to NTUFall 2027
Not announced yet
Computing and Data Science PhD
- Round
- Main
- Opens
- —
- GRE
- not required
- English
- TOEFL/IELTS
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- HKCUHK
Liwei Wang
- Vision-language & multimodal
- CHEPFL
Andrea Cavallaro
- Recognition, detection & segmentation
- CNTsinghua
Xueqian Wang
- Computational imaging & photography
- USUC Berkeley
Trevor Darrell
- Vision-language & multimodal
- CNZhejiang
Yao-Xiang Ding
- Reasoning & neuro-symbolic AI
- CNTsinghua
Liangrui Peng
- Interfaces & interaction techniques
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The lab site and the paper record do not agree.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.