影响力指数

74.39/100

前 2%

全站排名 #1,271

发表论文21 篇

平均评分5.3

年均产出7.0 篇/年

Xiaolin Huang

Full Professor@Shanghai Jiao Tong University·中国·OpenReview

研究方向

machine unlearning · optimization for deep neural networks · kernel learning · adversarial attack and defense · piecewise linear system

6.5

RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking Format

ICLR 2026Oral

通讯

6.0

Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models

ICLR 2026Poster

通讯

5.5

Pursuing Feature Separation based on Neural Collapse for Out-of-Distribution Detection

ICLR 2025Poster

通讯

6.6

Primphormer: Efficient Graph Transformers with Primal Representations

ICML 2025Poster

通讯

6.6

Flat-LoRA: Low-Rank Adaptation over a Flat Loss Landscape

ICML 2025Poster

通讯

6.3

Simulating Training Dynamics to Reconstruct Training Data from Deep Neural Networks

ICLR 2025Poster

通讯

5.8

Flat-LoRA: Low-Rank Adaption over a Flat Loss Landscape

ICLR 2025Rejected

通讯

5.0

Primphormer: Leveraging Primal Representation for Graph Transformers

合作者 (20)

Xiaolin Huang

RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking Format

Bi-LoRA: Efficient Sharpness-Aware Minimization for Fine-Tuning Large-Scale Models

T2I-ConBench: Text-to-Image Benchmark for Continual Post-training

A Practical Descent Method for Singular Value Decomposition

GradCFG: Gradient Inversion of Classifier-Free Guidance Diffusion Models

Zeroth-Order Forward-Only SNN Training Inspiring Neuromorphic On-Chip Learning

InjecMEM: Memory Injection Attack on LLM Agent Memory Systems