Dahua Lin
Associate Professor
Works on computer vision and machine learning, including large-scale visual understanding and generative models.
Dahua Lin's lab at CUHK works on recognition, detection & segmentation and generative & foundation models. 210 papers since 2023, 179 at top venues, mostly CVPR and ICLR. Applications close .
At a glance
- papers, 2023–2026
- 55/ year
- at T1 venues
- 85%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom DBLP, 2023–2026
- 40 papers2026 · T1 38 · collecting
- 60 papers2025 · T1 52
- 80 papers2024 · T1 63
- 30 papers2023 · T1 26
T1 venuesOtherVenues: CVPR, ICLR, NeurIPS, ICCV, Findings of ACL
Research focuspaper titles that use each topic's words
- Generative & foundation models42 papers
- Recognition, detection & segmentation9 papers
- large-scale visual0 papers
- visual understanding0 papers
- information engineering0 papers
Paper list
From DBLP.
202640 papers
- ICLR (T1)Achieving Olympia-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning (opens in a new tab)
- ICLR (T1)Advancing Complex Video Object Segmentation via Progressive Concept Construction (opens in a new tab)
- CVPR (T1)ARM-Thinker: Reinforcing Multimodal Generative Reward Models with Agentic Tool Use and Visual Reasoning (opens in a new tab)
- ICLR (T1)Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Models (opens in a new tab)
- ICLR (T1)CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning (opens in a new tab)
- CVPR (T1)ConsistCompose: Unified Multimodal Layout Control for Image Composition (opens in a new tab)
- ICLR (T1)DiCache: Let Diffusion Model Determine Its Own Cache (opens in a new tab)
- SIGGRAPH (T1)EAG-PT: Emission-Aware Gaussians and Path Tracing for Diffuse Indoor Scene Reconstruction and Editing (opens in a new tab)
- CVPR (T1)ELV-Halluc: Benchmarking Semantic Aggregation Hallucinations in Video Understanding (opens in a new tab)
- ECCV (T1)End-to-End Training for Autoregressive Video Diffusion via Self-resampling (opens in a new tab)
- ICLR (T1)FlexLinearAttention: Compiling a Unified Abstraction into Scalable Kernels for Linear Attention (opens in a new tab)
- ICLR (T1)From Pixels to Words -- Towards Native Vision-Language Primitives at Scale (opens in a new tab)
202560 papers
- CVPR (T1)3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion (opens in a new tab)
- ICLR (T1)3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation (opens in a new tab)
- Neural NetworksA survey of low-bit large language models: Basics, systems, and algorithms (opens in a new tab)
- TOG (T1)AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views (opens in a new tab)
- ICCV (T1)Bootstrap3D: Improving Multi-View Diffusion Model with Synthetic Data (opens in a new tab)
- CVPR (T1)ByTheWay: Boost Your Text-to-Video Generation Model to Higher Quality in a Training-free Way (opens in a new tab)
- COLING (T1)Case2Code: Scalable Synthetic Data for Code Generation (opens in a new tab)
- CVPR (T1)Conical Visual Concentration for Efficient Large Vision-Language Models (opens in a new tab)
- Findings of ACLConsultant Decoding: Yet Another Synergistic Mechanism (opens in a new tab)
- ICCV (T1)Creation-Mmbench: Assessing Context-Aware Creative Intelligence in Mllms (opens in a new tab)
- Findings of ACLdaDPO: Distribution-Aware DPO for Distilling Conversational Abilities (opens in a new tab)
- CVPR (T1)Dispider: Enabling Video LLMs with Active Real-Time Interaction via Disentangled Perception, Decision, and Reaction (opens in a new tab)
202480 papers
- DAC (T1)A Holistic Functionalization Approach to Optimizing Imperative Tensor Programs in Deep Learning (opens in a new tab)
- APNetA Knowledge-driven Self-healing Dual-loop and Validation for Autonomous Networks (opens in a new tab)
- NAACL (T1)Ada-LEval: Evaluating long-context LLMs with length-adaptable benchmarks (opens in a new tab)
- Findings of ACLAgent-FLAN: Designing Data and Methods of Effective Agent Tuning for Large Language Models (opens in a new tab)
- NeurIPS (T1)AlchemistCoder: Harmonizing and Eliciting Code Capability by Hindsight Tuning on Multi-source Data (opens in a new tab)
- CVPR (T1)Alpha-CLIP: A CLIP Model Focusing on Wherever you Want (opens in a new tab)
- NeurIPS (T1)ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models (opens in a new tab)
- ACL (T1)ANAH: Analytical Annotation of Hallucinations in Large Language Models (opens in a new tab)
- ICLR (T1)AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning (opens in a new tab)
- NeurIPS (T1)Are We on the Right Way for Evaluating Large Vision-Language Models? (opens in a new tab)
- Findings of ACLBalanced Data Sampling for Language Model Training with Clustering (opens in a new tab)
- ECCV (T1)Betrayed by Attention: A Simple yet Effective Approach for Self-supervised Video Object Segmentation (opens in a new tab)
202330 papers
- IEEE Trans. Multim.A Coarse-to-Fine Framework for Automatic Video Unscreen (opens in a new tab)
- ICCV (T1)AssetField: Assets Mining and Reconfiguration in Ground Feature Plane Representation (opens in a new tab)
- CVPR WorkshopCali-NCE: Boosting Cross-modal Video Representation Learning with Calibrated Alignment (opens in a new tab)
- HPCA (T1)Chimera: An Analytical Optimizing Framework for Effective Compute-intensive Operators Fusion (opens in a new tab)
- EMNLP (T1)CLEVA: Chinese Language Models EVAluation Platform (opens in a new tab)
- CVPR (T1)Controllable Mesh Generation Through Sparse Latent Point Diffusion Models (opens in a new tab)
- ICCV (T1)DNA-Rendering: A Diverse Neural Actor Repository for High-Fidelity Human-centric Rendering (opens in a new tab)
- CoRL (T1)DORT: Modeling Dynamic Objects in Recurrent for Multi-Camera 3D Object Detection and Tracking (opens in a new tab)
- SIGGRAPH PostersDynamic Storyboard Generation in an Engine-based Virtual Environment for Video Production (opens in a new tab)
- ICAIFE2EAI: End-to-End Deep Learning Framework for Active Investing (opens in a new tab)
- CVPR (T1)Grid-guided Neural Radiance Fields for Large Urban Scenes (opens in a new tab)
- IJCAI (T1)HireVAE: An Online and Adaptive Factor Model Based on Hierarchical and Regime-Switch VAE (opens in a new tab)
Where alumni went
Alumni are not listed on the lab homepage.
Apply to CUHKFall 2027
Computer Science and Engineering MPhil/PhD
- Round
- HKPFS / main round
- Opens
- —
- GRE
- See program page
- English
- Interview in English
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USStanford
Jiajun Wu
- Embodied & active vision
- USNYU
Yao Wang
- Video understanding
- CHEPFL
Haitham Al Hassanieh
- Computational imaging & photography
- GBImperial
Stefanos Zafeiriou
- 3D vision & neural rendering
- CAToronto
David Fleet
- 3D vision & neural rendering
- SGNTU
Shijian Lu
- Video understanding
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The member list could not be found on the lab site.
- The alumni list could not be found on the lab site.
- Collected
- Verified
- Papers
- DBLP
- Members
- Not collected
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.
CUHKcloses