Xia Hu

papers

1,087

total citations

papers (21)

A Comprehensive Study on Large-Scale Graph Training: Benchmarking and Rethinking

NEURIPS 2022arXiv

citations

DreamShard: Generalizable Embedding Table Placement for Recommender Systems

NEURIPS 2022arXiv

citations

Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts

CVPR 2024arXiv

citations

TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model

CVPR 2025arXiv

citations

DivAug: Plug-In Automated Data Augmentation With Explicit Diversity Maximization

ICCV 2021arXiv

citations

Setting the Trap: Capturing and Defeating Backdoors in Pretrained Language Models through Honeypots

NEURIPS 2023arXiv

citations

Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model

NEURIPS 2023arXiv

citations

Chasing Fairness Under Distribution Shift: A Model Weight Perturbation Approach

NEURIPS 2023arXiv

citations

MQuAKE-Remastered: Multi-Hop Knowledge Editing Can Only Be Advanced with Reliable Evaluations

ICLR 2025

citations

TVE: Learning Meta-attribution for Transferable Vision Explainer

ICML 2024arXiv

citations

Flexible Group Count Enables Hassle-Free Structured Pruning

CVPR 2025

citations

One Less Reason for Filter Pruning: Gaining Free Adversarial Robustness with Structured Grouped Kernel Pruning

NEURIPS 2023

citations

GNNs Also Deserve Editing, and They Need It More Than Once

ICML 2024

citations

Fair Graph Distillation

NEURIPS 2023

citations

Detecting Interactions from Neural Networks via Topological Analysis

NEURIPS 2020

citations

LLM Maybe LongLM: SelfExtend LLM Context Window Without Tuning

ICML 2024

citations

Soft Prompt Recovers Compressed LLMs, Transferably

ICML 2024

citations

Xia Hu

papers (21)

KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache

Towards Deeper Graph Neural Networks with Differentiable Group Normalization

Dirichlet Energy Constrained Learning for Deep Graph Neural Networks

Fairness via Representation Neutralization

A Comprehensive Study on Large-Scale Graph Training: Benchmarking and Rethinking

DreamShard: Generalizable Embedding Table Placement for Recommender Systems

Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts

TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model

DivAug: Plug-In Automated Data Augmentation With Explicit Diversity Maximization

Setting the Trap: Capturing and Defeating Backdoors in Pretrained Language Models through Honeypots

Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model

Chasing Fairness Under Distribution Shift: A Model Weight Perturbation Approach

MQuAKE-Remastered: Multi-Hop Knowledge Editing Can Only Be Advanced with Reliable Evaluations

TVE: Learning Meta-attribution for Transferable Vision Explainer

Flexible Group Count Enables Hassle-Free Structured Pruning

One Less Reason for Filter Pruning: Gaining Free Adversarial Robustness with Structured Grouped Kernel Pruning

GNNs Also Deserve Editing, and They Need It More Than Once

Fair Graph Distillation

Detecting Interactions from Neural Networks via Topological Analysis

LLM Maybe LongLM: SelfExtend LLM Context Window Without Tuning

Soft Prompt Recovers Compressed LLMs, Transferably

papers (21)

KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache

Towards Deeper Graph Neural Networks with Differentiable Group Normalization

Dirichlet Energy Constrained Learning for Deep Graph Neural Networks

Fairness via Representation Neutralization

A Comprehensive Study on Large-Scale Graph Training: Benchmarking and Rethinking

DreamShard: Generalizable Embedding Table Placement for Recommender Systems

Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts

TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model

DivAug: Plug-In Automated Data Augmentation With Explicit Diversity Maximization

Setting the Trap: Capturing and Defeating Backdoors in Pretrained Language Models through Honeypots

Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model

Chasing Fairness Under Distribution Shift: A Model Weight Perturbation Approach

MQuAKE-Remastered: Multi-Hop Knowledge Editing Can Only Be Advanced with Reliable Evaluations

TVE: Learning Meta-attribution for Transferable Vision Explainer

Flexible Group Count Enables Hassle-Free Structured Pruning

One Less Reason for Filter Pruning: Gaining Free Adversarial Robustness with Structured Grouped Kernel Pruning

GNNs Also Deserve Editing, and They Need It More Than Once

Fair Graph Distillation

Detecting Interactions from Neural Networks via Topological Analysis

LLM Maybe LongLM: SelfExtend LLM Context Window Without Tuning

Soft Prompt Recovers Compressed LLMs, Transferably