Yifei Wang

Affiliations

MIT CSAIL

papers

648

total citations

papers (22)

How Mask Matters: Towards Theoretical Understandings of Masked Autoencoders

NEURIPS 2022arXiv

citations

Dissecting the Diffusion Process in Linear Graph Convolutional Networks

NEURIPS 2021arXiv

citations

When Adversarial Training Meets Vision Transformers: Recipes from Training to Architecture

NEURIPS 2022arXiv

citations

Improving Out-of-Distribution Generalization by Adversarial Training with Structured Priors

NEURIPS 2022arXiv

citations

On the Duality Between Sharpness-Aware Minimization and Adversarial Training

ICML 2024arXiv

citations

Laplacian Canonization: A Minimalist Approach to Sign and Basis Invariant Spectral Embedding

NEURIPS 2023arXiv

citations

Adversarial Examples Are Not Real Features

NEURIPS 2023arXiv

citations

Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspective

NEURIPS 2023arXiv

citations

Identifiable Contrastive Learning with Automatic Feature Importance Discovery

NEURIPS 2023arXiv

citations

Architecture Matters: Uncovering Implicit Mechanisms in Graph Contrastive Learning

NEURIPS 2023arXiv

citations

OODRobustBench: a Benchmark and Large-Scale Analysis of Adversarial Robustness under Distribution Shift

ICML 2024arXiv

citations

Look Ahead or Look Around? A Theoretical Comparison Between Autoregressive and Masked Pretraining

ICML 2024arXiv

citations

Uni-Instruct: One-step Diffusion Model through Unified Diffusion Divergence Instruction

NEURIPS 2025arXiv

citations

Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness

ICLR 2025arXiv

citations

$\texttt{G1}$: Teaching LLMs to Reason on Graphs with Reinforcement Learning

NEURIPS 2025

citations

Semi-IIN: Semi-Supervised Intra-Inter Modal Interaction Learning Network for Multimodal Sentiment Analysis

AAAI 2025arXiv

citations

Dynamic Context-Sensitive Filtering Network for Video Salient Object Detection

ICCV 2021

citations

Beyond the Best: Distribution Functional Estimation in Infinite-Armed Bandits

NEURIPS 2022

citations

Yifei Wang

Affiliations

papers (22)

How Mask Matters: Towards Theoretical Understandings of Masked Autoencoders

Dissecting the Diffusion Process in Linear Graph Convolutional Networks

When Adversarial Training Meets Vision Transformers: Recipes from Training to Architecture

CFA: Class-Wise Calibrated Fair Adversarial Training

Residual Relaxation for Multi-view Representation Learning

Do Generated Data Always Help Contrastive Learning?

On the Emergence of Position Bias in Transformers

Improving Out-of-Distribution Generalization by Adversarial Training with Structured Priors

On the Duality Between Sharpness-Aware Minimization and Adversarial Training

Laplacian Canonization: A Minimalist Approach to Sign and Basis Invariant Spectral Embedding

Adversarial Examples Are Not Real Features

Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspective

Identifiable Contrastive Learning with Automatic Feature Importance Discovery

Architecture Matters: Uncovering Implicit Mechanisms in Graph Contrastive Learning

OODRobustBench: a Benchmark and Large-Scale Analysis of Adversarial Robustness under Distribution Shift

Look Ahead or Look Around? A Theoretical Comparison Between Autoregressive and Masked Pretraining

Uni-Instruct: One-step Diffusion Model through Unified Diffusion Divergence Instruction

Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness

$\texttt{G1}$: Teaching LLMs to Reason on Graphs with Reinforcement Learning

Semi-IIN: Semi-Supervised Intra-Inter Modal Interaction Learning Network for Multimodal Sentiment Analysis

Dynamic Context-Sensitive Filtering Network for Video Salient Object Detection

Beyond the Best: Distribution Functional Estimation in Infinite-Armed Bandits

papers (22)

How Mask Matters: Towards Theoretical Understandings of Masked Autoencoders

Dissecting the Diffusion Process in Linear Graph Convolutional Networks

When Adversarial Training Meets Vision Transformers: Recipes from Training to Architecture

CFA: Class-Wise Calibrated Fair Adversarial Training

Residual Relaxation for Multi-view Representation Learning

Do Generated Data Always Help Contrastive Learning?

On the Emergence of Position Bias in Transformers

Improving Out-of-Distribution Generalization by Adversarial Training with Structured Priors

On the Duality Between Sharpness-Aware Minimization and Adversarial Training

Laplacian Canonization: A Minimalist Approach to Sign and Basis Invariant Spectral Embedding

Adversarial Examples Are Not Real Features

Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspective

Identifiable Contrastive Learning with Automatic Feature Importance Discovery

Architecture Matters: Uncovering Implicit Mechanisms in Graph Contrastive Learning

OODRobustBench: a Benchmark and Large-Scale Analysis of Adversarial Robustness under Distribution Shift

Look Ahead or Look Around? A Theoretical Comparison Between Autoregressive and Masked Pretraining

Uni-Instruct: One-step Diffusion Model through Unified Diffusion Divergence Instruction

Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness

$\texttt{G1}$: Teaching LLMs to Reason on Graphs with Reinforcement Learning

Semi-IIN: Semi-Supervised Intra-Inter Modal Interaction Learning Network for Multimodal Sentiment Analysis

Dynamic Context-Sensitive Filtering Network for Video Salient Object Detection

Beyond the Best: Distribution Functional Estimation in Infinite-Armed Bandits