Chun Yuan
Professor
Works on computer vision, multimedia, and machine learning, including video object segmentation, detector distillation, and compatible representation learning.
Chun Yuan's lab at Tsinghua works on recognition, detection & segmentation, video understanding and generative & foundation models. 129 papers since 2023, 82 at top venues, mostly CVPR and AAAI. Applications close .
At a glance
- papers, 2023–2026
- 34/ year
- at T1 venues
- 64%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom DBLP, 2023–2026
- 19 papers2026 · T1 15 · collecting
- 40 papers2025 · T1 29
- 38 papers2024 · T1 18
- 32 papers2023 · T1 20
T1 venuesOtherVenues: CVPR, AAAI, ICCV, ICML, ICLR
Research focuspaper titles that use each topic's words
- Recognition, detection & segmentation20 papers
- Generative & foundation models16 papers
- distillation8 papers
- Video understanding7 papers
- segmentation5 papers
Paper list
From DBLP.
202619 papers
- Findings of ACLDiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding (opens in a new tab)
- Findings of ACLEfficient Transformer Parameter Reuse via Zero-Token Mechanism (opens in a new tab)
- ICLR (T1)Enhancing Geometric Perception in VLMs via Translator-Guided Reinforcement Learning (opens in a new tab)
- CVPR (T1)FlashVSR: Towards Real-time Diffusion-Based Streaming Video Super Resolution (opens in a new tab)
- IEEE Trans. Multim.High-Frequency Prioritized Sparse Attention Network for Image Restoration (opens in a new tab)
- CVPR (T1)Learning Cross-View Object Correspondence via Cycle-Consistent Mask Prediction (opens in a new tab)
- CVPR (T1)Love Me, Love My Label: Rethinking the Role of Labels in Prompt Retrieval for Visual In-Context Learning (opens in a new tab)
- AAAI (T1)M3Time: LLM-Enhanced Multi-Modal, Multi-Scale, and Multi-Frequency Multivariate Time Series Forecasting (opens in a new tab)
- ICLR (T1)OmniPortrait: Fine-Grained Personalized Portrait Synthesis via Pivotal Optimization (opens in a new tab)
- ICLR (T1)OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging (opens in a new tab)
- ACL (T1)PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering (opens in a new tab)
- ICLR (T1)PromptHub: Enhancing Multi-Prompt Visual In-Context Learning with Locality-Aware Fusion, Concentration and Alignment (opens in a new tab)
202540 papers
- NeurIPS (T1)A Simple Linear Patch Revives Layer-Pruned Large Language Models (opens in a new tab)
- TPAMI (T1)Accelerating Zero-Shot NAS With Feature Map-Based Proxy and Operation Scoring Function (opens in a new tab)
- AAAI (T1)Aligning Composed Query with Image via Discriminative Perception from Negative Correspondences (opens in a new tab)
- IEEE Trans. Multim.Boosting Long-Tailed Recognition With Label Descriptor and Beyond (opens in a new tab)
- ICLR (T1)ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding (opens in a new tab)
- ICCV (T1)ChartPoint: Guiding MLLMs with Grounding Reflection for Chart Reasoning (opens in a new tab)
- ICIPClip-Ae: Clip-Assisted Cross-View Audio-Visual Enhancement for Unsupervised Temporal Action Localization (opens in a new tab)
- SIGGRAPH (T1)Cobra: Efficient Line Art COlorization with BRoAder References (opens in a new tab)
- CVPR (T1)ComRoPE: Scalable and Robust Rotary Position Embedding Parameterized by Trainable Commuting Angle Matrices (opens in a new tab)
- ICMEEAV-Mamba: Efficient Audio-Visual Representation Learning for Weakly-Supervised Temporal Action Localization (opens in a new tab)
- IEEE Trans. Circuits Syst. Video Technol.EDTformer: An Efficient Decoder Transformer for Visual Place Recognition (opens in a new tab)
- ICML (T1)Enhancing Logits Distillation with Plug&Play Kendall's τ Ranking Loss (opens in a new tab)
202438 papers
- ECCV (T1)A Task Is Worth One Word: Learning with Task Prompts for High-Quality Versatile Image Inpainting (opens in a new tab)
- ICPRAMC-OA: Adaptive Multi-Scale Convolutional Networks with Optimized Attention for Temporal Action Localization (opens in a new tab)
- ICPRBenchmarking AI in Mental Health: A Critical Examination of LLMs Across Key Performance and Ethical Metrics (opens in a new tab)
- AAAI (T1)Blind Face Restoration under Extreme Conditions: Leveraging 3D-2D Prior Fusion for Superior Structural and Texture Recovery (opens in a new tab)
- ECCV WorkshopBoosting Pose Estimators via Cross-Representation Distillation (opens in a new tab)
- ICLR (T1)Convolution Meets LoRA: Parameter Efficient Finetuning for Segment Anything Model (opens in a new tab)
- CVPR (T1)CricaVPR: Cross-Image Correlation-Aware Representation Learning for Visual Place Recognition (opens in a new tab)
- ICML (T1)CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers (opens in a new tab)
- ACM MMCustomNet: Object Customization with Variable-Viewpoints in Text-to-Image Diffusion Models (opens in a new tab)
- AAAI (T1)Deep Homography Estimation for Visual Place Recognition (opens in a new tab)
- ICML (T1)DFD: Distilling the Feature Disparity Differently for Detectors (opens in a new tab)
- CVPR (T1)Distilling Semantic Priors from SAM to Efficient Image Restoration Models (opens in a new tab)
202332 papers
- ICRA (T1)AANet: Aggregation and Alignment Network with Semi-hard Positive Sample Mining for Hierarchical Place Recognition (opens in a new tab)
- ICCV (T1)Accurate 3D Face Reconstruction with Facial Component Tokens (opens in a new tab)
- ACM MMAdaptive Contrastive Learning for Learning Robust Representations under Label Noise (opens in a new tab)
- CVPR (T1)Architecture, Dataset and Model-Scale Agnostic Data-free Meta-Learning (opens in a new tab)
- AAAI (T1)Darwinian Model Upgrades: Model Evolving with Selective Compatibility (opens in a new tab)
- Big DataDesign of Fatigue Driving Behavior Detection Based on Circle Hough Transform (opens in a new tab)
- IJCAI (T1)DFVSR: Directional Frequency Video Super-Resolution via Asymmetric and Enhancement Alignment Network (opens in a new tab)
- ICCV WorkshopEffective Whole-body Pose Estimation with Two-stages Distillation (opens in a new tab)
- Journal of Artificial Intelligence ResearchEfficient Multi-Goal Reinforcement Learning via Value Consistency Prioritization (opens in a new tab)
- ACM MMEnhanced Image Deblurring: An Efficient Frequency Exploitation and Preservation Network (opens in a new tab)
- ICASSP (T1)Frequency Reciprocal Action and Fusion for Single Image Super-Resolution (opens in a new tab)
- ICCV (T1)From Knowledge Distillation to Self-Knowledge Distillation: A Unified Approach with Normalized Loss and Customized Soft Labels (opens in a new tab)
Where alumni went
Alumni are not listed on the lab homepage.
Apply to TsinghuaFall 2027
Computer Science PhD (international applicants)
- Round
- Main
- Opens
- GRE
- See program page
- English
- Per school catalog
- Fee
- RMB 800
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USMIT
William Freeman
- Video understanding
- GBImperial
Maja Pantic
- Recognition, detection & segmentation
- CNTsinghua
Xiangyang Ji
- Computational imaging & photography
- USCMU
Alexander Hauptmann
- Video understanding
- USUT Austin
Chen Yu
- Video understanding
- CNPeking
Chao Xu
- 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.
Tsinghuacloses