Hongxu Yin

papers

3,218

total citations

papers (20)

Recurrence Without Recurrence: Stable Video Landmark Detection With Deep Equilibrium Models

CVPR 2023arXiv

citations

Adaptive Sharpness-Aware Pruning for Robust Sparse Networks

ICLR 2024arXiv

citations

LANA: Latency Aware Network Acceleration

ECCV 2022arXiv

citations

Heterogeneous Continual Learning

CVPR 2023arXiv

citations

LLaMaFlex: Many-in-one LLMs via Generalized Pruning and Weight Sharing

ICLR 2025

citations

Optimal Quantization Using Scaled Codebook

CVPR 2021

citations

Token-Efficient VLM: High-Resolution Image Understanding via Dynamic Region Proposal

ICCV 2025

citations

A-ViT: Adaptive Tokens for Efficient Vision Transformer

CVPR 2022

citations

Hongxu Yin

papers (20)

DoRA: Weight-Decomposed Low-Rank Adaptation

VILA: On Pre-training for Visual Language Models

Dreaming to Distill: Data-Free Knowledge Transfer via DeepInversion

See Through Gradients: Image Batch Recovery via GradInversion

FasterViT: Fast Vision Transformers with Hierarchical Attention

GradViT: Gradient Inversion of Vision Transformers

Global Vision Transformer Pruning With Hessian-Aware Saliency

When To Prune? A Policy Towards Early Structural Pruning

Structural Pruning via Latency-Saliency Knapsack

WorldModelBench: Judging Video Generation Models As World Models

FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Models

Flextron: Many-in-One Flexible Large Language Model

Recurrence Without Recurrence: Stable Video Landmark Detection With Deep Equilibrium Models

Adaptive Sharpness-Aware Pruning for Robust Sparse Networks

LANA: Latency Aware Network Acceleration

Heterogeneous Continual Learning

LLaMaFlex: Many-in-one LLMs via Generalized Pruning and Weight Sharing

Optimal Quantization Using Scaled Codebook

Token-Efficient VLM: High-Resolution Image Understanding via Dynamic Region Proposal

A-ViT: Adaptive Tokens for Efficient Vision Transformer

papers (20)

DoRA: Weight-Decomposed Low-Rank Adaptation

VILA: On Pre-training for Visual Language Models

Dreaming to Distill: Data-Free Knowledge Transfer via DeepInversion

See Through Gradients: Image Batch Recovery via GradInversion

FasterViT: Fast Vision Transformers with Hierarchical Attention

GradViT: Gradient Inversion of Vision Transformers

Global Vision Transformer Pruning With Hessian-Aware Saliency

When To Prune? A Policy Towards Early Structural Pruning

Structural Pruning via Latency-Saliency Knapsack

WorldModelBench: Judging Video Generation Models As World Models

FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Models

Flextron: Many-in-One Flexible Large Language Model

Recurrence Without Recurrence: Stable Video Landmark Detection With Deep Equilibrium Models

Adaptive Sharpness-Aware Pruning for Robust Sparse Networks

LANA: Latency Aware Network Acceleration

Heterogeneous Continual Learning

LLaMaFlex: Many-in-one LLMs via Generalized Pruning and Weight Sharing

Optimal Quantization Using Scaled Codebook

Token-Efficient VLM: High-Resolution Image Understanding via Dynamic Region Proposal

A-ViT: Adaptive Tokens for Efficient Vision Transformer