影响力指数

61.92/100

前 4.3%

全站排名 #2,779

发表论文14 篇

平均评分5.4

年均产出4.7 篇/年

Peiran Yu

Assistant Professor@University of Texas at Arlington·美国·OpenReview

研究方向

large langage model · minimax optimization · bilevel optimization · adversarial training · hyperparameter optimization · continuous Optimization

5.3

Zeroth-Order Methods for Stochastic Nonconvex Nonsmooth Composite Optimization

ICLR 2026Rejected

二作

5.3

New Hybrid Fine-Tuning Paradigm for LLMs: Algorithm Design and Convergence Analysis Framework

ICLR 2026Poster

二作

4.5

Mitigating Data Heterogeneity Effect in Client-Reshuffling-Based Federated Learning

ICLR 2026Rejected

二作

4.5

Breaking the Correlation Plateau: On the Optimization and Capacity Limits of Attention-Based Regressors

ICLR 2026Poster

三作

7.6

Provably Mitigating Corruption, Overoptimization, and Verbosity Simultaneously in Offline and Online RLHF/DPO Alignment

ICLR 2025Rejected

三作

5.5

Bilevel ZOFO: Bridging Parameter-Efficient and Zeroth-Order Techniques for Efficient LLM Fine-Tuning and Meta-Training

ICLR 2025Rejected

三作

4.0

A Fast Federated Method for Minimax Problems with Sequential Convergence Guarantees

ICLR 2025Withdrawn

一作

4.0

Hybrid Fine-Tuning of LLMs: Theoretical Insights on Generalized Smoothness and Convergence

ICLR 2025Withdrawn

二作

3.8

Provably Mitigating Corruption, Overoptimization, and Verbosity Simultaneously in Offline and Online RLHF/DPO Alignment

合作者 (12)

Peiran Yu

Zeroth-Order Methods for Stochastic Nonconvex Nonsmooth Composite Optimization

New Hybrid Fine-Tuning Paradigm for LLMs: Algorithm Design and Convergence Analysis Framework

Mitigating Data Heterogeneity Effect in Client-Reshuffling-Based Federated Learning

Breaking the Correlation Plateau: On the Optimization and Capacity Limits of Attention-Based Regressors

Cost-Aware Contrastive Routing for LLMs

Bilevel ZOFO: Efficient LLM Fine-Tuning and Meta-Training

Revisiting Convergence: Shuffling Complexity Beyond Lipschitz Smoothness

Revisiting Convergence: A Study on Shuffling-Type Gradient Methods

Provably Mitigating Corruption, Overoptimization, and Verbosity Simultaneously in Offline and Online RLHF/DPO Alignment

Bilevel ZOFO: Bridging Parameter-Efficient and Zeroth-Order Techniques for Efficient LLM Fine-Tuning and Meta-Training

A Fast Federated Method for Minimax Problems with Sequential Convergence Guarantees

Hybrid Fine-Tuning of LLMs: Theoretical Insights on Generalized Smoothness and Convergence

Provably Mitigating Corruption, Overoptimization, and Verbosity Simultaneously in Offline and Online RLHF/DPO Alignment