Profiles2021-2024
Xinmeng Huang
University of Pennsylvania
10Papers
0Last author
16Unique co-authors
5First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers10
- Momentum Benefits Non-iid Federated Learning Simply and ProvablyAuthor 2 of 4ICLR 2024 · 40 citations
- One-Shot Safety Alignment for Large Language Models via Optimal DualizationAuthor 1 of 6NeurIPS 2024 · 30 citations
- Uncertainty in Language Models: Assessment through Rank-CalibrationAuthor 1 of 8EMNLP 2024 · 9 citations
- SPARKLE: A Unified Single-Loop Primal-Dual Framework for Decentralized Bilevel OptimizationAuthor 4 of 6NeurIPS 2024 · 7 citations
- Unbiased Compression Saves Communication in Distributed Optimization: When and How Much?Author 2 of 3NeurIPS 2023 · 25 citations
- Demystifying Disagreement-on-the-Line in High DimensionsAuthor 3 of 5ICML 2023 · 12 citations
- Lower Bounds and Nearly Optimal Algorithms in Distributed Learning with Communication CompressionAuthor 1 of 4NeurIPS 2022 · 49 citations
- Revisiting Optimal Convergence Rate for Smooth and Non-convex Stochastic Decentralized OptimizationAuthor 2 of 6NeurIPS 2022 · 40 citations
- Collaborative Learning of Discrete Distributions under Heterogeneity and Communication ConstraintsAuthor 1 of 4NeurIPS 2022 · 3 citations
- An Improved Analysis and Rates for Variance Reduction under Without-replacement Sampling OrdersAuthor 1 of 4NeurIPS 2021 · 1 citation
