publications

2025

  1. COLING
    QuickLLaMA: Query-aware Inference Acceleration for Large Language Models
    Jingyao Li, Han Shi, Xin Jiang, and 3 more authors
    In International Conference on Computational Linguistics, 2025

2024

  1. arXiv
    SepLLM: Accelerate Large Language Models by Compressing One Segment into One Token
    Guoxuan Chen, Han Shi‡, Jiawei Li, and 7 more authors
    Preprint arXiv:2412.12094, 2024
  2. NeurIPS
    DAPE: Data-Adaptive Positional Encoding for Length Extrapolation
    Chuanyang Zheng, Yihang Gao, Han Shi, and 8 more authors
    In Neural Information Processing Systems, 2024
  3. NeurIPS
    Diffusion of Thoughts: Chain-of-Thought Reasoning in Diffusion Language Models
    Jiacheng Ye, Shansan Gong, Liheng Chen, and 8 more authors
    In Neural Information Processing Systems, 2024
  4. ICLRSpotlight
    MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
    Longhui Yu, Weisen Jiang, Han Shi‡, and 7 more authors
    In International Conference on Learning Representations, 2024
  5. ICLROral
    LEGO-Prover: Neural Theorem Proving with Growing Libraries
    Haiming Wang, Huajian Xin, Chuanyang Zheng, and 8 more authors
    In International Conference on Learning Representations, 2024
  6. Findings of ACL
    Forward-Backward Reasoning in Large Language Models for Mathematical Verification
    Weisen Jiang, Han Shi, Longhui Yu, and 4 more authors
    In Findings of the Association for Computational Linguistics, 2024
  7. arXiv
    Accelerating Auto-regressive Text-to-Image Generation with Training-free Speculative Jacobi Decoding
    Yao Teng, Han Shi, Xian Liu, and 5 more authors
    Preprint arXiv:2410.01699, 2024
  8. arXiv
    Efficient Multi-modal Large Language Models via Visual Token Grouping
    Minbin Huang, Runhui Huang, Han Shi, and 6 more authors
    Preprint arXiv:2411.17773, 2024
  9. arXiv
    DAPE V2: Process Attention Score as Feature Map for Length Extrapolation
    Chuanyang Zheng, Yihang Gao, Han Shi, and 8 more authors
    preprint arXiv:2410.04798, 2024
  10. arXiv
    DiM: Diffusion Mamba for Efficient High-Resolution Image Synthesis
    Yao Teng, Yue Wu, Han Shi, and 5 more authors
    Preprint arXiv:2405.14224, 2024
  11. arXiv
    On the Expressive Power of a Variant of the Looped Transformer
    Yihang Gao, Chuanyang Zheng, Enze Xie, and 6 more authors
    Preprint arXiv:2402.13572, 2024

2023

  1. ACL
    DT-Solver: Automated Theorem Proving with Dynamic-Tree Sampling Guided by Proof-level Value Function
    Haiming Wang, Ye Yuan, Zhengying Liu, and 8 more authors
    In Association for Computational Linguistics, 2023
  2. ICCVOral
    DiffFit: Unlocking Transferability of Large Diffusion Models via Simple Parameter-Efficient Fine-Tuning
    Enze Xie, Lewei Yao, Han Shi, and 5 more authors
    In International Conference on Computer Vision, 2023
  3. ICCV
    GrowCLIP: Data-aware Automatic Model Growing for Large-scale Contrastive Language-Image Pre-training
    Xinchi Deng, Han Shi, Runhui Huang, and 7 more authors
    In International Conference on Computer Vision, 2023

2022

  1. CVPR
    Continual Object Detection via Prototypical Task Correlation Guided Gating Mechanism
    Binbin Yang, Xinchi Deng, Han Shi, and 6 more authors
    In Computer Vision and Pattern Recognition, 2022
  2. ICLR
    Revisiting Over-smoothing in BERT from the Perspective of Graph
    Han Shi, Jiahui Gao, Hang Xu, and 5 more authors
    In International Conference on Learning Representations, 2022
  3. AAAI
    AutoBERT-Zero: Evolving BERT Backbone from Scratch
    Jiahui Gao, Hang Xu, Han Shi, and 5 more authors
    In AAAI Conference on Artificial Intelligence, 2022

2021

  1. ICML
    SparseBERT: Rethinking the Importance Analysis in Self-attention
    Han Shi, Jiahui Gao, Xiaozhe Ren, and 4 more authors
    In International Conference on Machine Learning, 2021

2020

  1. NeurIPS
    Bridging the Gap between Sample-based and One-shot Neural Architecture Search with BONAS
    Han Shi, Renjie Pi, Hang Xu, and 3 more authors
    Neural Information Processing Systems, 2020
  2. AAAI
    Effective Decoding in Graph Auto-Encoder using Triadic Closure
    Han Shi, Haozheng Fan, and James T Kwok
    In AAAI Conference on Artificial Intelligence, 2020