publications

Publications by categories in reversed chronological order.

2026

  1. ICML
    How Does Reasoning Flow? Tracing Attention-Induced Information Flow for Targeted RL in LLMs
    Zhichen Dong*, Yang Li*, Yuhan Sun, and 7 more authors
    In International Conference on Machine Learning, 2026
  2. ICML
    Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization
    Yang Li, Zhichen Dong, Yuhan Sun, and 9 more authors
    In International Conference on Machine Learning, 2026
  3. CVPR
    Reasoning Palette: Modulating Reasoning via Latent Contextualization for Controllable Exploration for (V)LMs
    Rujiao Long*, Yang Li*, Xingyao Zhang, and 7 more authors
    In Conference on Computer Vision and Pattern Recognition, 2026
  4. ICLR
    MaskCO: Masked Generation Drives Effective Representation Learning and Exploiting for Combinatorial Optimization
    Lvda Chen*, Yang Li*, and Junchi Yan
    In International Conference on Learning Representations, 2026
  5. ICML
    Problem Distributions as Tasks: Repurposing Meta Learning for Generative Combinatorial Optimization towards Multi-task Pretrain and Adaptation
    Wenzheng Pan, Jiale Ma, Nuoyan Chen, and 2 more authors
    In International Conference on Machine Learning, 2026
  6. ICLR
    ConRep4CO: Contrastive Representation Learning of Combinatorial Optimization Instances across Types
    Ziao Guo, Yang Li, Shiyue Wang, and 1 more author
    In International Conference on Learning Representations, 2026
  7. ICLR
    Native Adaptive Solution Expansion for Diffusion-based Combinatorial Optimization
    Yu Wang, Yang Li, Jiale Ma, and 2 more authors
    In International Conference on Learning Representations, 2026
  8. ICLR
    Smarter Not Harder: Generative Process Evaluation with Intrinsic-Signal Driving and Ability-Adaptive Reward Shaping
    Tao He, Rongchuan Mu, Lizi Liao, and 6 more authors
    In International Conference on Learning Representations, 2026
  9. arXiv
    Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem
    ROCK and ROLL and IFLOW and DT Joint Team
    2026
    Core contributor

2025

  1. arXiv
    Part II: ROLL Flash–Accelerating RLVR and Agentic Training with Asynchrony
    Han Lu, Zichen Liu, Shaopan Xiong, and 19 more authors
    2025
  2. NeurIPS
    Bridging Crypto with ML-based Solvers: the SAT Formulation and Benchmarks
    Xinhao Zheng, Xinhao Song, Bolin Qiu, and 3 more authors
    In Advances in Neural Information Processing Systems, 2025
  3. NeurIPS
    Generation as Search Operator for Test-Time Scaling of Diffusion-based Combinatorial Optimization
    Yang Li, Lvda Chen, Haonan Wang, and 2 more authors
    In Advances in Neural Information Processing Systems, 2025
  4. NeurIPS
    StruDiCO: Structured Denoising Diffusion with Gradient-free Inference-stage Boosting for Memory and Time Efficient Combinatorial Optimization
    Yu Wang, Yang Li, Junchi Yan, and 1 more author
    In Advances in Neural Information Processing Systems, 2025
  5. NeurIPS
    ML4CO-Bench-101: Benchmark Machine Learning for Classic Combinatorial Problems on Graphs
    Jiale Ma, Wenzheng Pan, Yang Li, and 1 more author
    In The Thirty-ninth Annual Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2025
  6. ICML
    Generative Modeling Reinvents Supervised Learning: Label Repurposing with Predictive Consistency Learning
    Yang Li, Jiale Ma, Yebin Yang, and 3 more authors
    In International Conference on Machine Learning, 2025
  7. ICML
    COExpander: Adaptive Solution Expansion for Combinatorial Optimization
    Jiale Ma, Wenzheng Pan, Yang Li, and 1 more author
    In International Conference on Machine Learning, 2025
  8. ICLR
    Unify ML4TSP: Drawing Methodological Principles for TSP and Beyond from Streamlined Design Space of Learning and Search
    Yang Li, Jiale Ma, Wenzheng Pan, and 4 more authors
    In International Conference on Learning Representations, 2025
  9. ICLR
    UniCO: On Unified Combinatorial Optimization via Problem Reduction to Matrix-Encoded General TSP
    Wenzheng Pan, Hao Xiong, Jiale Ma, and 3 more authors
    In International Conference on Learning Representations, 2025

2024

  1. NeurIPS
    Fast T2T: Optimization Consistency Speeds Up Diffusion-Based Training-to-Testing Solving for Combinatorial Optimization
    Yang Li*, Jinpei Guo*, Runzhong Wang, and 2 more authors
    In Advances in Neural Information Processing Systems, 2024
  2. NeurIPS
    Learning Plaintext-Ciphertext Cryptographic Problems via ANF-based SAT Instance Representation
    Xinhao Zheng, Yang Li, Cunxin Fan, and 3 more authors
    In Advances in Neural Information Processing Systems, 2024
  3. NeurIPS
    Benchmarking PtO and PnO Methods in the Predictive Combinatorial Optimization Regime
    Haoyu Geng, Hang Ruan, Runzhong Wang, and 4 more authors
    In Advances in Neural Information Processing Systems, 2024
  4. 中国科学
    Learning to Solve Combinatorial Optimization under Positive Linear Constraints via Non-Autoregressive Neural Networks
    Runzhong Wang, Yang Li, Junchi Yan, and 1 more author
    SCIENTIA SINICA Informationis 2024, 2024
  5. ICMLSpotlight (3.5%)
    ACM-MILP: Adaptive Constraint Modification via Grouping and Selection for Hardness-Preserving MILP Instance Generation
    Ziao Guo, Yang Li, Chang Liu, and 2 more authors
    In The Forty-first International Conference on Machine Learning, 2024
  6. ICLR
    MixSATGEN: Learning Graph Mixing for SAT Instance Generation
    Xinyan Chen*, Yang Li*, Runzhong Wang, and 1 more author
    In The Twelfth International Conference on Learning Representations, 2024
  7. arXiv
    Machine Learning Insides OptVerse AI Solver: Design Principles and Applications
    Xijun Li, Fangzhou Zhu, Hui-Ling Zhen, and 23 more authors
    arXiv preprint, 2024
  8. arXiv
    Molecule Generation for Drug Design: a Graph Learning Perspective
    Nianzu Yang, Huaijin Wu, Kaipeng Zeng, and 2 more authors
    In Fundamental Research, 2024

2023

  1. NeurIPS
    T2T: From Distribution Learning in Training to Gradient Search in Testing for Combinatorial Optimization
    Yang Li, Jinpei Guo, Runzhong Wang, and 1 more author
    In Advances in Neural Information Processing Systems, 2023
  2. SIGKDD
    HardSATGEN: Understanding the Difficulty of Hard SAT Formula Generation and A Strong Structure-Hardness-Aware Baseline
    Yang Li, Xinyan Chen, Wenxuan Guo, and 6 more authors
    In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2023
  3. IJCAI
    IID-GAN: an IID Sampling Perspective for Regularizing Mode Collapse
    Yang Li*, Liangliang Shi*, and Junchi Yan
    In Proceedings of the 32nd International Joint Conference on Artificial Intelligence, 2023

2022

  1. NeurIPSSpotlight (5%)
    Improving Generative Adversarial Networks via Adversarial Learning in Latent Space
    Yang Li, Yichuan Mo, Liangliang Shi, and 1 more author
    In Advances in Neural Information Processing Systems, 2022
  2. NeurIPS
    The Policy-gradient Placement and Generative Routing Neural Networks for Chip Design
    Ruoyu Cheng, Xianglong Lyu, Yang Li, and 3 more authors
    In Advances in Neural Information Processing Systems, 2022
  3. SAT Comp.
    Kissat Adaptive Restart, Kissat Cfexp: Adaptive Restart Policy and Variable Scoring Improvement
    Yang Li, Yuqi Jia, Wanqian Luo, and 4 more authors
    Proceedings of SAT Competition, 2022
  4. SAT Comp.
    The SAT Encoding for Graph Isomorphism
    Yang Li, Yuqi Jia, Wanqian Luo, and 4 more authors
    Proceedings of SAT Competition, 2022