Publications

Publications by categories in reversed chronological order. Also see this Google Scholar page or this Semantic Scholar page. The * denotes equal contribution.


Preprints

2024

  1. Preprint
    CRISPR-GPT: LLM Agents for Automated Design of Gene-Editing Experiments
    Yuanhao Qu*, Kaixuan Huang*, Ming Yin, Kanghong Zhan, Dyllan Liu, Di Yin, William A Johnson, Xiaotong Wang, Denny Zhou, Russ Altman, Mengdi Wang*, and Le Cong*
    Under Review by Nature Biomedical Engineering, 2024
  2. Preprint
    Relative-Translation Invariant Wasserstein Distance
    Binshuai Wang, Qiwei Di, Ming Yin, Mengdi Wang, Quanquan Gu, and Peng Wei
    arXiv preprint arXiv:2409.02416, 2024
    arXiv
  3. Preprint
    Posterior Sampling via Langevin Monte Carlo for Offline Reinforcement Learning
    Thanh Nguyen-Tang, Ming Yin, Masatoshi Uehara, Yu-Xiang Wang, Mengdi Wang, and Raman Arora
    2024
    HTML


Publications

2024

  1. NeurIPS
    A Theoretical Perspective for Speculative Decoding Algorithm
    Ming Yin, Minshuo Chen, Kaixuan Huang, and Mengdi Wang
    Advances in Neural Information Processing Systems, 2024
    arXiv
  2. NeurIPS
    NetworkGym: Reinforcement Learning Environments for Multi-Access Traffic Management in Network Simulation
    Momin Haider, Ming Yin, Menglei Zhang, Arpit Gupta, Jing Zhu, and Yu-Xiang Wang
    Advances in Neural Information Processing Systems (Datasets and Benchmarks Track), 2024
    arXiv
  3. NeurIPS
    Fast Inference-Time Alignment via Speculative Rejection
    Ruiqi Zhang*, Momin Haider*, Ming Yin, Jiahao Qiu, Mengdi Wang, Peter Bartlett, and Andrea Zanette
    Advances in Neural Information Processing Systems, 2024
    arXiv
  4. NeurIPS
    Transfer Q*: Principled Decoding for LLM Alignment
    Souradip Chakraborty, Soumya Ghosal, Ming Yin, Dinesh Manocha, Mengdi Wang, Amrit Singh Bedi, and Furong Huang
    Advances in Neural Information Processing Systems, 2024
    arXiv
  5. NeurIPS
    Offline Multitask Representation Learning for Reinforcement Learning
    Haque Ishfaq*, Thanh Nguyen-Tang, Songtao Feng, Raman Arora, Mengdi Wang, Ming Yin*, and Doina Precup*
    Advances in Neural Information Processing Systems, 2024
    arXiv
  6. ICML
    Learning the Target Network in Function Space
    Ming Yin*, Kavosh Asadi*, Shoham Sabach*, Yao Liu*, and Rasool Fakoor
    International Conference on Machine Learning, 2024
    arXiv HTML
  7. ICML
    Improving Sample Efficiency of Model-Free Algorithms for Zero-Sum Markov Games
    Songtao Feng, Ming Yin, Yu-Xiang Wang, Jing Yang, and Yingbin Liang
    International Conference on Machine Learning, 2024
    arXiv HTML
  8. JSAIT
    Towards General Function Approximation in Nonstationary Reinforcement Learning
    Songtao Feng, Ming Yin, Ruiquan Huang, Yu-Xiang Wang, Jing Yang, and Yingbin Liang
    IEEE Journal on Selected Areas in Information Theory, (short version at ISIT 24) 2024
    HTML
  9. CVPR Best Paper Finalist
    MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
    Xiang Yue, Yuansheng Ni, Kai Zhang, Tianyu Zheng, Ruoqi Liu, Ge Zhang, Samuel Stevens, Dongfu Jiang, Weiming Ren, Yuxuan Sun, Cong Wei, Botao Yu, Ruibin Yuan, Renliang Sun, Ming Yin, Boyuan Zheng, Zhenzhu Yang, Yibo Liu, Wenhao Huang, Huan Sun, Yu Su, and Wenhu Chen
    The IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024
    arXiv HTML
  10. ISIT
    Logarithmic Switching Cost in Reinforcement Learning beyond Linear MDPs
    Dan Qiao, Ming Yin, and Yu-Xiang Wang
    IEEE International Symposium on Information Theory 2024
    arXiv

2023

  1. NeurIPS
    Posterior Sampling with Delayed Feedback for Reinforcement Learning with Linear Function Approximation
    Ming Yin*, Nikki Kuang*, Mengdi Wang, Yu-Xiang Wang, and Yian Ma
    Advances in Neural Information Processing Systems, 2023
    arXiv HTML
  2. EMNLP
    TheoremQA: A Theorem-driven Question Answering dataset
    Wenhu Chen, Ming Yin, Max Ku, Elaine Wan, Xueguang Ma, Jianyu Xu, Tony Xia, Xinyi Wang, and Pan Lu
    Conference on Empirical Methods in Natural Language Processing, 2023
    arXiv HTML
  3. UAI
    No-Regret Linear Bandits beyond Realizability
    Chong Liu, Ming Yin, and Yu-Xiang Wang
    Uncertainty in Artificial Intelligence, 2023
    arXiv HTML
  4. ICML
    Non-stationary Reinforcement Learning under General Function Approximation
    Songtao Feng, Ming Yin, Ruiquan Huang, Yu-Xiang Wang, Jing Yang, and Yingbin Liang
    International Conference on Machine Learning, 2023
    arXiv HTML
  5. ICML
    Offline Reinforcement Learning with Closed-Form Policy Improvement Operators
    Jiachen Li, Edwin Zhang, Ming Yin, Qinxun Bai, Yu-Xiang Wang, and William Yang Wang
    International Conference on Machine Learning (Short version at NeurIPS-22 Offline RL Workshop), 2023
    arXiv HTML
  6. ICLR
    Offline Reinforcement Learning with Differentiable Function Approximation is Provably Efficient
    Ming Yin, Mengdi Wang, and Yu-Xiang Wang
    International Conference on Learning Representations, 2023
    arXiv HTML
  7. AAAI
    On Instance-Dependent Bounds for Offline Reinforcement Learning with Linear Function Approximation
    Thanh Nguyen-Tan, Ming Yin, Sunil Gupta, Svetha Venkates, and Raman Arora
    Association for the Advancement of Artificial Intelligence, 2023
    arXiv HTML

2022

  1. ICLR
    Near-optimal Offline Reinforcement Learning with Linear Representation: Leveraging Variance Information with Pessimism
    Ming Yin, Yaqi Duan, Mengdi Wang, and Yu-Xiang Wang
    International Conference on Learning Representations, 2022
    arXiv HTML
  2. ICML Spotlight
    Sample-Efficient Reinforcement Learning with loglog(T) Switching Cost
    Dan Qiao, Ming Yin, Ming Min, and Yu-Xiang Wang
    International Conference on Machine Learning, 2022
    arXiv HTML
  3. UAI Spotlight
    Offline Stochastic Shortest Path: Learning, Evaluation and Towards Optimality
    Ming Yin*, Wenjing Chen*, Mengdi Wang, and Yu-Xiang Wang
    Uncertainty in Artificial Intelligence, 2022
    arXiv HTML

2021

  1. NeurIPS
    Towards Instance-optimal Offline Reinforcement Learning with Pessimism
    Ming Yin, and Yu-Xiang Wang
    Advances in Neural Information Processing Systems, 2021
    arXiv HTML
  2. NeurIPS
    Optimal Uniform OPE and Model-based Offline Reinforcement Learning in Time-Homogeneous, Reward-Free and Task-Agnostic Settings
    Ming Yin, and Yu-Xiang Wang
    Advances in Neural Information Processing Systems (Short version at ICML RL Theory Workshop), 2021
    arXiv HTML
  3. NeurIPS
    Near-Optimal Offline Reinforcement Learning via Double Variance Reduction
    Ming Yin, Yu Bai, and Yu-Xiang Wang
    Advances in Neural Information Processing Systems (Short version at ICML RL Theory Workshop), 2021
    arXiv HTML
  4. AISTATS Oral Presentation
    Near-optimal Provable Uniform Convergence in Offline Policy Evaluation for Reinforcement Learning
    Ming Yin, Yu Bai, and Yu-Xiang Wang
    International Conference on Artificial Intelligence and Statistics (Short version at Neurips 2020 Offline RL Workshop), 2021
    arXiv HTML

2020

  1. AISTATS
    Asymptotically Efficient Off-policy Evaluation for Tabular Reinforcement Learning
    Ming Yin, and Yu-Xiang Wang
    International Conference on Artificial Intelligence and Statistics, 2020
    arXiv HTML

Workshop Papers

2023

  1. ICML-WS
    Why Quantization Improves Generalization: NTK of Binary Weight Neural Networks
    Kaiqi Zhang, Ming Yin, and Yu-Xiang Wang
    ICML workshop in Neural Compression, 2023
    arXiv

    2022

      1. NeurIPS-WS
        Offline Policy Evaluation for Reinforcement Learning with Adaptively Collected Data
        Sunil Madhow, Dan Qiao, Ming Yin, and Yu-Xiang Wang
        NeurIPS workshop in Offline RL, 2022
        HTML