Zhouhan Lin
教授
Works on machine learning and natural language processing, including language model architectures.
Zhouhan Lin's lab at SJTU works on large language models and accelerators, chips & architecture. 18 papers since 2023, 17 at top venues, mostly ICLR and ICML. Applications close .
At a glance
- papers, 2023–2026
- 5/ year
- at T1 venues
- 94%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom the lab homepage, 2023–2026
- 7 papers2026 · T1 7 · collecting
- 6 papers2025 · T1 5
- 3 papers2024 · T1 3
- 2 papers2023 · T1 2
T1 venuesOtherVenues: ICLR, ICML, NeurIPS, ACL, ACM MM
Research focuspaper titles that use each topic's words
- Large language models9 papers
- language7 papers
- Accelerators, chips & architecture0 papers
- model architectures0 papers
- language model0 papers
Paper list
From the lab homepage.
20267 papers
- ICLR (T1)AWM: Accurate Weight-Matrix Fingerprint for Large Language Models
- ICML (T1)Context-level Language Modeling by Learning Predictive Context Embeddings
- ICLR (T1)Flow of Spans: Generalizing Language Models to Dynamic Span-Vocabulary via GFlowNets
- ECCV (T1)Fourier Compressor: Frequency-Domain Visual Token Compression for Vision-Language Models
- ICLR (T1)FreqKV: Key-Value Compression in Frequency Domain for Context Window Extension
- ICML (T1)PonderLM-2: Pretraining LLM with Latent Thoughts in Continuous Space
- ICLR (T1)PonderLM: Pretraining Language Models to Ponder in Continuous Space
20256 papers
- ICML (T1)AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence
- ACL (T1)Gumbel Reranking: Differentiable End-to-End Reranker Optimization
- ICML (T1)How to Synthesize Text Data without Model Collapse?
- NeurIPS (T1)Memory Decoder: A Pretrained, Plug-and-Play Memory for Large Language Models
- EMNLP (T1)Training LLMs to be Better Text Embedders through Bidirectional Reconstruction
- ACM MMVisual-informed Silent Video Identity Conversion
20243 papers
- NeurIPS (T1)Cluster-wise Graph Transformer with Dual-granularity Kernelized Attention
- ICLR (T1)Graph Parsing Networks
- NeurIPS (T1)HuRef: HUman-REadable Fingerprint for Large Language Models
20232 papers
- ICLR (T1)Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothing
- NeurIPS (T1)Tailoring Self-Attention for Graph via Rooted Subtrees
Where alumni went
Alumni are not listed on the lab homepage.
Apply to SJTUFall 2027
Computer Science PhD (international applicants)
- Round
- Self-funded
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USCMU
Jamie Callan
- Knowledge, QA & retrieval
- USUIUC
Romit Roy Choudhury
- Signal processing & information theory
- USCMU
Emma Strubell
- Large language models
- CNTsinghua
Zaiqing Nie
- Large language models
- CNTsinghua
Bin Xu
- Large language models
- USUCLA
Yangruibo Ding
- Large language models
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
- The lab homepage
- Members
- Not collected
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.
SJTUcloses