Changshui Zhang
Professor
Works on pattern recognition, machine learning, and computer vision, including few-shot learning, causal learning, and reasoning with large models.
Changshui Zhang's lab at Tsinghua works on probabilistic & causal inference, recognition, detection & segmentation and large language models. 28 papers since 2023, 21 at top venues, mostly ICML and ICLR. 11 PhD students and 6 master's students. Applications close .
At a glance
- papers, 2023–2026
- 7/ year
- at T1 venues
- 75%
- PhD students
- 11
- MS students
- 6
- alumni listed
- —
Papers by yearfrom the lab homepage, 2023–2026
- 4 papers2026 · T1 4 · collecting
- 5 papers2025 · T1 4
- 12 papers2024 · T1 10
- 7 papers2023 · T1 3
T1 venuesOtherVenues: ICML, ICLR, ACL, ICRA, Pattern Recognition
Research focuspaper titles that use each topic's words
- Large language models9 papers
- Probabilistic & causal inference0 papers
- Recognition, detection & segmentation0 papers
- few-shot learning0 papers
- few-shot0 papers
Paper list
From the lab homepage.
20264 papers
- ICLR (T1)Attend to the Active: Structure-Aware Dynamic Attention in LLMs for Compositional Instruction Following
- AAAI (T1)Beyond Tokens: Dynamic Latent Reasoning via Semantic Residual Refinement
- ICRA (T1)Gold Points Sniper: Empowering VLMs with Self-Guided Reasoning for Visually-Grounded Action Understanding
- ICRA (T1)Learning Push-Grasp Synergy for Occluded Objects in Cluttered Environments
20255 papers
- ICML (T1)Adaptive Localization of Knowledge Negation for Continual LLM Unlearning
- Findings of ACLAssimilation and Accommodation: Task-Adaptive Hierarchical Abstraction for Solving Web Tasks
- ICML (T1)CALM: Consensus-Aware Localized Merging for Multi-Task Learning
- ICML (T1)Learning without Isolation: Pathway Protection for Continual Learning
- COLING (T1)Physics Reasoner: Knowledge-Augmented Reasoning for Solving Physics Problems with Large Language Models
202412 papers
- ICML (T1)2024
- NAACL (T1)A Closer Look at the Self-Verification Abilities of Large Language Models in Logical Reasoning
- TPAMI (T1)A Theoretical View of Linear Backpropagation and Its Convergence
- In Proceedings of the Conference of the Empirical Methods in Natural Language ProcessingAbstraction-of-Thought Makes Language Model Better Reasoner
- ICLR (T1)Accurate Forgetting for Heterogeneous Federated Continual Learning
- ICLR (T1)CLAP: Collaborative Adaptation for Checkerboard Learning
- ACL (T1)CLoMo: Counterfactual Logical Modification with Large Language Models
- NeurocopumtingFrom Text to Mask: Localizing Entities Using the Attention of Text-to-Image Diffusion Models
- ICASSP (T1)Learned ISTA with Error-based Thresholding for Adaptive Sparse Coding
- TNNLS (T1)Meta-Modulation: A General Learning Framework for Cross-Task Adaptation
- NeurIPS (T1)Neural Collapse Inspired Feature Alignment for Out-of-Distribution Generalization
- EMNLP (T1)Subjective Topic meets LLMs: Unleashing Comprehensive, Reflective and Creative Thinking through the Negation of Negation
20237 papers
- Pattern RecognitionA Survey on Machine Learning from Few Samples
- Cybernetics and Intelligence. Tsinghua University PressAdvances in Reasoning by Prompting Large Language Models: A Survey
- Journal of Chemical Information and ModelingAutomatically Predicting Material Properties with Microscopic Images: Polymer Miscibility as an Example
- TPAMI (T1)Bipartite Ranking Fairness through a Model Agnostic Ordering Adjustment
- TNNLS (T1)CATRO: Channel Pruning via Class-Aware Trace Ratio Optimization
- ACL (T1)Faithful Question Answering with Monte-Carlo Planning
- Pattern RecognitionWhere You Edit is What You Get: Text-Guided Image Editing with Region-Based Attention
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
- CNSJTU
Dequan Wang
13 T1 - SGNTU
Daniel Paulin
- Probabilistic & causal inference
- CAToronto
Daniel Roy
- Learning theory & optimization
- CNZhejiang
Jun Xiao
60 T1 - USUMD
Yaodong Yu
- Generative & foundation models
- USMIT
Tomaso Poggio
- Computational neuroscience & cognition
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The alumni list could not be found on the lab site.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.
Tsinghuacloses