Li Dong

Affiliations

Microsoft Research

papers

9,947

total citations

papers (21)

Swin Transformer V2: Scaling Up Capacity and Resolution

CVPR 2022arXiv

2,487

citations

Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks

ECCV 2020arXiv

2,159

citations

MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers

NEURIPS 2020arXiv

1,843

citations

Extensible Prompts for Language Models on Zero-shot Language Style Customization

NEURIPS 2023arXiv

citations

Learning Robust Image Watermarking with Lossless Cover Recovery

ICCV 2025

citations

Image as a Foreign Language: BEiT Pretraining for Vision and Vision-Language Tasks

CVPR 2023

citations

Differential Transformer

ICLR 2025arXiv

citations

Semi-Parametric Retrieval via Binary Bag-of-Tokens Index

ICLR 2025arXiv

citations

Li Dong

Affiliations

papers (21)

Swin Transformer V2: Scaling Up Capacity and Resolution

Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks

MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers

Grounding Multimodal Large Language Models to the World

VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts

Language Is Not All You Need: Aligning Perception with Language Models

Optimizing Prompts for Text-to-Image Generation

BioCLIP: A Vision Foundation Model for the Tree of Life

Augmenting Language Models with Long-Term Memory

On the Representation Collapse of Sparse Mixture of Experts

Imagine While Reasoning in Space: Multimodal Visualization-of-Thought

Think Only When You Need with Large Hybrid-Reasoning Models

Scaling Laws of Synthetic Data for Language Model

Non-Contrastive Learning Meets Language-Image Pre-Training

Self-Boosting Large Language Models with Synthetic Preference Data

Generic-to-Specific Distillation of Masked Autoencoders

Extensible Prompts for Language Models on Zero-shot Language Style Customization

Learning Robust Image Watermarking with Lossless Cover Recovery

Image as a Foreign Language: BEiT Pretraining for Vision and Vision-Language Tasks

Differential Transformer

Semi-Parametric Retrieval via Binary Bag-of-Tokens Index

papers (21)

Swin Transformer V2: Scaling Up Capacity and Resolution

Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks

MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers

Grounding Multimodal Large Language Models to the World

VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts

Language Is Not All You Need: Aligning Perception with Language Models

Optimizing Prompts for Text-to-Image Generation

BioCLIP: A Vision Foundation Model for the Tree of Life

Augmenting Language Models with Long-Term Memory

On the Representation Collapse of Sparse Mixture of Experts

Imagine While Reasoning in Space: Multimodal Visualization-of-Thought

Think Only When You Need with Large Hybrid-Reasoning Models

Scaling Laws of Synthetic Data for Language Model

Non-Contrastive Learning Meets Language-Image Pre-Training

Self-Boosting Large Language Models with Synthetic Preference Data

Generic-to-Specific Distillation of Masked Autoencoders

Extensible Prompts for Language Models on Zero-shot Language Style Customization

Learning Robust Image Watermarking with Lossless Cover Recovery

Image as a Foreign Language: BEiT Pretraining for Vision and Vision-Language Tasks

Differential Transformer

Semi-Parametric Retrieval via Binary Bag-of-Tokens Index