[
    {
        "title": "Settling the Burn-in Gap: Sharp Risk Bounds for Tail-Averaged SGD in Linear Regression",
        "authors": "Qingyue Zhao*, Xuheng Li*, Quanquan Gu",
        "type": "preprint",
        "research_area": "optimization",
        "year": 2026,
        "pdf": "https://arxiv.org/pdf/2611.13383",
        "arxiv": "https://arxiv.org/abs/2611.13383"
    },
    {
        "title": "High-Probability Regret Bounds for Feel-Good Thompson Sampling",
        "authors": "Xuheng Li, Qingyue Zhao, Quanquan Gu",
        "type": "preprint",
        "research_area": "rl-llm",
        "year": 2026,
        "pdf": "https://arxiv.org/pdf/2610.19513",
        "arxiv": "https://arxiv.org/abs/2610.19513",
        "image": "img/High-Probability Regret Bounds.png"
    },
    {
        "title": "Muon Learns Facts Better: Understanding the Role of Spectral Orthogonalization",
        "authors": "Xuheng Li, Qiwei Di, Yuan Cao, Quanquan Gu",
        "type": "preprint",
        "research_area": "optimization",
        "year": 2026,
        "pdf": "https://arxiv.org/pdf/2610.02798",
        "arxiv": "https://arxiv.org/abs/2610.02798",
        "image": "img/Muon Learns Facts Better.svg"
    },
    {
        "title": "Query-Efficient Rejection Sampling for Parallel Masked Diffusion",
        "authors": "Shiyuan Zhang*, Xuheng Li*, Qiwei Di, Quanquan Gu",
        "type": "preprint",
        "research_area": "sampling",
        "year": 2026,
        "pdf": "https://arxiv.org/pdf/2610.22238",
        "arxiv": "https://arxiv.org/abs/2610.22238"
    },
    {
        "title": "Understanding Off-vs On-Policy Distillation: A Tale of Distinct Training Objectives",
        "authors": "Qiwei Di, Xuheng Li, Kaixuan Ji, Chenggong Zhang, Heyang Zhao, Quanquan Gu",
        "type": "preprint",
        "research_area": "rl-llm",
        "year": 2026,
        "pdf": "https://arxiv.org/pdf/2609.38666",
        "arxiv": "https://arxiv.org/abs/2609.38666",
        "image": "img/Understanding Off-vs On-Policy Distillation.svg"
    },
    {
        "title": "Saturation-Insensitive Dueling Bandits with General Function Approximation",
        "authors": "Chenggong Zhang, Xuheng Li, Qiwei Di, Weitong Zhang, Quanquan Gu",
        "type": "preprint",
        "research_area": "rl-llm",
        "year": 2026,
        "pdf": "https://arxiv.org/pdf/2609.33011",
        "arxiv": "https://arxiv.org/abs/2609.33011",
        "image": "img/Saturation-Insensitive Dueling Bandits.svg"
    },
    {
        "title": "Dimension-Independent Convergence of Underdamped Langevin Monte Carlo in KL Divergence",
        "authors": "Shiyuan Zhang*, Qiwei Di*, Xuheng Li*, Quanquan Gu",
        "type": "conference",
        "research_area": "sampling",
        "year": 2026,
        "conference_name": "the 43rd International Conference on Machine Learning (ICML)",
        "conference_location": "Seoul, South Korea",
        "pdf": "https://openreview.net/pdf?id=gxxOL0iXpr",
        "openreview": "https://openreview.net/forum?id=gxxOL0iXpr",
        "arxiv": "https://arxiv.org/abs/2603.02429",
        "image": "img/Dimension-Independent.png"
    },
    {
        "title": "Best-of-Majority: Minimax-Optimal Strategy for Pass@<em>k</em> Inference Scaling",
        "authors": "Qiwei Di*, Kaixuan Ji*, Xuheng Li*, Heyang Zhao, Quanquan Gu",
        "type": "conference",
        "research_area": "rl-llm",
        "year": 2026,
        "conference_name": "the 14th International Conference on Learning Representations (ICLR)",
        "pdf": "https://openreview.net/pdf?id=a6CVQpjbXq",
        "openreview": "https://openreview.net/forum?id=a6CVQpjbXq",
        "conference_location": "Rio de Janeiro, Brazil",
        "arxiv": "https://arxiv.org/abs/2510.03199",
        "image": "img/Best-of-Majority.png"
    },
    {
        "title": "Understanding SGD with Exponential Moving Average: A Case Study in Linear Regression",
        "authors": "Xuheng Li, Quanquan Gu",
        "type": "preprint",
        "research_area": "optimization",
        "year": 2025,
        "pdf": "https://arxiv.org/pdf/2502.14123",
        "arxiv": "https://arxiv.org/abs/2502.14123",
        "image": "img/Understanding SGD.png"
    },
    {
        "title": "Variance-Aware Feel-Good Thompson Sampling for Contextual Bandits",
        "authors": "Xuheng Li, Quanquan Gu",
        "type": "conference",
        "research_area": "rl-llm",
        "year": 2025,
        "conference_name": "the 39th Conference on Neural Information Processing Systems (NeurIPS)",
        "pdf": "https://openreview.net/pdf?id=eGpI9rz1Pw",
        "openreview": "https://openreview.net/forum?id=eGpI9rz1Pw",
        "conference_location": "San Diego, California, USA",
        "arxiv": "https://arxiv.org/abs/2511.02123",
        "image": "img/Variance-Aware Feel-Good.png"
    },
    {
        "title": "Feel-Good Thompson Sampling for Contextual Dueling Bandits",
        "authors": "Xuheng Li, Heyang Zhao, Quanquan Gu",
        "type": "conference",
        "research_area": "rl-llm",
        "year": 2024,
        "conference_name": "the 41st International Conference on Machine Learning (ICML)",
        "pdf": "https://openreview.net/pdf?id=l9ga3iQuHt",
        "openreview": "https://openreview.net/forum?id=l9ga3iQuHt",
        "conference_location": "Vienna, Austria",
        "arxiv": "https://arxiv.org/abs/2404.06013",
        "image": "img/Feel-Good Thompson.png"
    },
    {
        "title": "Risk Bounds of Accelerated SGD for Overparameterized Linear Regression",
        "authors": "Xuheng Li, Yihe Deng, Jingfeng Wu, Dongruo Zhou, Quanquan Gu",
        "type": "conference",
        "research_area": "optimization",
        "year": 2024,
        "conference_name": "the 12th International Conference on Learning Representations (ICLR)",
        "pdf": "https://openreview.net/pdf?id=AcoXPIPh4A",
        "openreview": "https://openreview.net/forum?id=AcoXPIPh4A",
        "conference_location": "Vienna, Austria",
        "arxiv": "https://arxiv.org/abs/2311.14222",
        "image": "img/Risk Bounds of Accelerated SGD.png"
    }
]
