Yao Lu

papers

2,089

total citations

papers (20)

Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Research

NEURIPS 2023arXiv

163

citations

NVILA: Efficient Frontier Visual Language Models

CVPR 2025arXiv

157

citations

Grounded Decoding: Guiding Text Generation with Grounded Models for Embodied Agents

NEURIPS 2023arXiv

citations

Learning Optical Flow From a Few Matches

CVPR 2021arXiv

citations

RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models

CVPR 2025arXiv

citations

Scaling RL to Long Videos

NEURIPS 2025arXiv

citations

WorldModelBench: Judging Video Generation Models As World Models

NEURIPS 2025arXiv

citations

VILA-M3: Enhancing Vision-Language Models with Medical Expert Knowledge

CVPR 2025arXiv

citations

Taskology: Utilizing Task Relations at Scale

CVPR 2021arXiv

citations

Token Turing Machines

CVPR 2023arXiv

citations

Understanding the Dynamics of DNNs Using Graph Modularity

ECCV 2022arXiv

citations

Scaling Vision Pre-Training to 4K Resolution

CVPR 2025arXiv

citations

DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer

ICCV 2025arXiv

citations

SparseVILA: Decoupling Visual Sparsity for Efficient VLM Inference

ICCV 2025arXiv

citations

Delayed Gradient Averaging: Tolerate the Communication Latency for Federated Learning

NEURIPS 2021

citations

ALRMR-GEC: Adjusting Learning Rate Based on Memory Rate to Optimize the Edit Scorer for Grammatical Error Correction

AAAI 2025

citations

A Set of Generalized Components to Achieve Effective Poison-only Clean-label Backdoor Attacks with Collaborative Sample Selection and Triggers

NEURIPS 2025arXiv

citations

Yao Lu

papers (20)

VILA: On Pre-training for Visual Language Models

Learning To Estimate Hidden Motions With Global Motion Aggregation

CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models

Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Research

NVILA: Efficient Frontier Visual Language Models

Grounded Decoding: Guiding Text Generation with Grounded Models for Embodied Agents

Learning Optical Flow From a Few Matches

RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models

Scaling RL to Long Videos

WorldModelBench: Judging Video Generation Models As World Models

VILA-M3: Enhancing Vision-Language Models with Medical Expert Knowledge

Taskology: Utilizing Task Relations at Scale

Token Turing Machines

Understanding the Dynamics of DNNs Using Graph Modularity

Scaling Vision Pre-Training to 4K Resolution

DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer

SparseVILA: Decoupling Visual Sparsity for Efficient VLM Inference

Delayed Gradient Averaging: Tolerate the Communication Latency for Federated Learning

ALRMR-GEC: Adjusting Learning Rate Based on Memory Rate to Optimize the Edit Scorer for Grammatical Error Correction

A Set of Generalized Components to Achieve Effective Poison-only Clean-label Backdoor Attacks with Collaborative Sample Selection and Triggers

papers (20)

VILA: On Pre-training for Visual Language Models

Learning To Estimate Hidden Motions With Global Motion Aggregation

CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models

Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Research

NVILA: Efficient Frontier Visual Language Models

Grounded Decoding: Guiding Text Generation with Grounded Models for Embodied Agents

Learning Optical Flow From a Few Matches

RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models

Scaling RL to Long Videos

WorldModelBench: Judging Video Generation Models As World Models

VILA-M3: Enhancing Vision-Language Models with Medical Expert Knowledge

Taskology: Utilizing Task Relations at Scale

Token Turing Machines

Understanding the Dynamics of DNNs Using Graph Modularity

Scaling Vision Pre-Training to 4K Resolution

DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer

SparseVILA: Decoupling Visual Sparsity for Efficient VLM Inference

Delayed Gradient Averaging: Tolerate the Communication Latency for Federated Learning

ALRMR-GEC: Adjusting Learning Rate Based on Memory Rate to Optimize the Edit Scorer for Grammatical Error Correction

A Set of Generalized Components to Achieve Effective Poison-only Clean-label Backdoor Attacks with Collaborative Sample Selection and Triggers