Hongxu Yin

18

Papers

720

Total Citations

Papers (18)

VILA: On Pre-training for Visual Language Models

WorldModelBench: Judging Video Generation Models As World Models

NeurIPS 2025arXiv

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

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

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

DoRA: Weight-Decomposed Low-Rank Adaptation

Flextron: Many-in-One Flexible Large Language Model

Dreaming to Distill: Data-Free Knowledge Transfer via DeepInversion

Optimal Quantization Using Scaled Codebook

See Through Gradients: Image Batch Recovery via GradInversion

GradViT: Gradient Inversion of Vision Transformers

When To Prune? A Policy Towards Early Structural Pruning

A-ViT: Adaptive Tokens for Efficient Vision Transformer

Global Vision Transformer Pruning With Hessian-Aware Saliency

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

Heterogeneous Continual Learning

LANA: Latency Aware Network Acceleration

Structural Pruning via Latency-Saliency Knapsack

NeurIPS 2022arXiv