Kai Han

Venues

Frequent coauthors

Papers

1. MATCH: Modulating Attention via In-Context Retrieval for Long-Context Transformers
   ACL 2026 ·

   Linrui Ma

   DBLP profile ↗ORCID search ↗
2. PocketLLM: Ultimate Compression of Large Language Models via Meta Networks
   AAAI 2026 ·

   Ye Tian

   DBLP profile ↗ORCID search ↗
3. DenseSSM: State Space Models with Dense Hidden Connection for Efficient Large Language Models
   NAACL 2025 ·

   Wei He

   DBLP profile ↗ORCID search ↗
4. EMS-SD: Efficient Multi-sample Speculative Decoding for Accelerating Large Language Models
   NAACL 2025 ·

   Yunsheng Ni

   DBLP profile ↗ORCID search ↗
5. Eve: Efficient Multimodal Vision Language Models with Elastic Visual Experts
   AAAI 2025 ·

   Miao Rang

   DBLP profile ↗ORCID search ↗
6. Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM Reasoning
   ICML 2025 ·

   Zhenni Bi

   DBLP profile ↗ORCID search ↗
7. GPT4Image: Large Pre-trained Models Help Vision Models Learn Better on Perception Task
   WWW 2025 ·

   Ning Ding

   DBLP profile ↗ORCID search ↗
8. L-Man: A Large Multi-modal Model Unifying Human-centric Tasks
   AAAI 2025 ·

   Jialong Zuo

   DBLP profile ↗ORCID search ↗
9. LLM Data Selection and Utilization via Dynamic Bi-level Optimization
   ICML 2025 ·

   Yang Yu

   DBLP profile ↗ORCID search ↗
10. Mixture of Lookup Experts
    ICML 2025 ·

    Shibo Jie

    DBLP profile ↗ORCID search ↗
11. SpeCache: Speculative Key-Value Caching for Efficient Generation of LLMs
    ICML 2025 ·

    Shibo Jie

    DBLP profile ↗ORCID search ↗
12. Adapt Without Forgetting: Distill Proximity from Dual Teachers in Vision-Language Models
    ECCV 2024 ·

    Mengyu Zheng

    DBLP profile ↗ORCID search ↗
13. An Empirical Study of Scaling Law for Scene Text Recognition
    CVPR 2024 ·

    Miao Rang

    DBLP profile ↗ORCID search ↗
14. Data-efficient Large Vision Models through Sequential Autoregression
    ICML 2024 ·

    Zhiwei Hao

    DBLP profile ↗ORCID search ↗
15. GeminiFusion: Efficient Pixel-wise Multimodal Fusion for Vision Transformer
    ICML 2024 ·

    Ding Jia

    DBLP profile ↗ORCID search ↗
16. Kangaroo: Lossless Self-Speculative Decoding for Accelerating LLMs via Double Early Exiting
    NeurIPS 2024 ·

    Fangcheng Liu

    DBLP profile ↗ORCID search ↗
17. Memory-Space Visual Prompting for Efficient Vision-Language Fine-Tuning
    ICML 2024 ·

    Shibo Jie

    DBLP profile ↗ORCID search ↗
18. MemoryFormer : Minimize Transformer Computation by Removing Fully-Connected Layers
    NeurIPS 2024 ·

    Ning Ding

    DBLP profile ↗ORCID search ↗
19. ParameterNet: Parameters are All You Need for Large-Scale Visual Pretraining of Mobile Networks
    CVPR 2024 · Kai Han
20. Rethinking Optimization and Architecture for Tiny Language Models
    ICML 2024 ·

    Yehui Tang

    DBLP profile ↗ORCID search ↗
21. Star-Agents: Automatic Data Optimization with LLM Agents for Instruction Tuning
    NeurIPS 2024 ·

    Hang Zhou

    DBLP profile ↗ORCID search ↗
22. Token Compensator: Altering Inference Cost of Vision Transformer Without Re-tuning
    ECCV 2024 ·

    Shibo Jie

    DBLP profile ↗ORCID search ↗