Profiles2021-2026
Haishan Ye
Xi'an Jiaotong University
13Papers
1Last author
16Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers13
- FZOO: Fast Zeroth-Order Optimizer for Fine‑Tuning Large Language Models towards Adam‑Scale SpeedAuthor 4 of 7ICLR 2026 · 16 citations
- MSCR: Exploring the Vulnerability of LLMs’ Mathematical Reasoning Abilities Using Multi-Source Candidate ReplacementAuthor 3 of 3ICLR 2026 · 1 citation
- A Near-Optimal Algorithm for Decentralized Convex-Concave Finite-Sum Minimax OptimizationAuthor 3 of 4NeurIPS 2025 · 2 citations
- ProAdvPrompter: A Two-Stage Journey to Effective Adversarial Prompting for LLMsAuthor 3 of 7ICLR 2025
- Second-Order Fine-Tuning without Pain for LLMs: A Hessian Informed Zeroth-Order OptimizerAuthor 3 of 6ICLR 2025
- Decentralized Riemannian Conjugate Gradient Method on the Stiefel ManifoldAuthor 2 of 7ICLR 2024 · 21 citations
- Double Stochasticity Gazes Faster: Snap-Shot Decentralized Stochastic Gradient Tracking MethodsAuthor 2 of 5ICML 2024 · 4 citations
- Double Variance Reduction: A Smoothing Trick for Composite Optimization Problems without First-Order GradientAuthor 2 of 6ICML 2024 · 2 citations
- Near-Optimal Distributed Minimax Optimization under the Second-Order SimilarityAuthor 2 of 3NeurIPS 2024 · 2 citations
- Can Gaussian Sketching Converge Faster on a Preconditioned Landscape?Author 2 of 4ICML 2024 · 1 citation
- Stochastic Distributed Optimization under Average Second-order Similarity: Algorithms and AnalysisAuthor 3 of 4NeurIPS 2023 · 17 citations
- Eigencurve: Optimal Learning Rate Schedule for SGD on Quadratic Objectives with Skewed Hessian SpectrumsAuthor 2 of 3ICLR 2022 · 20 citations
- Greedy and Random Quasi-Newton Methods with Faster Explicit Superlinear ConvergenceAuthor 2 of 3NeurIPS 2021 · 16 citations
