Hongyi Wang

papers

878

total citations

papers (10)

Attack of the Tails: Yes, You Really Can Backdoor Federated Learning

NEURIPS 2020arXiv

753

citations

Rare Gems: Finding Lottery Tickets at Initialization

NEURIPS 2022arXiv

citations

AMP: Automatically Finding Model Parallel Strategies with Heterogeneity Awareness

NEURIPS 2022arXiv

citations

CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing

COLM 2025arXiv

citations

M2OST: Many-to-one Regression for Predicting Spatial Transcriptomics from Digital Pathology Images

AAAI 2025arXiv

citations

FedNAR: Federated Optimization with Normalized Annealing Regularization

NEURIPS 2023arXiv

citations

Understanding the Forgetting of (Replay-based) Continual Learning via Feature Learning: Angle Matters

ICML 2025

citations

Hongyi Wang

papers (10)

Attack of the Tails: Yes, You Really Can Backdoor Federated Learning

Rare Gems: Finding Lottery Tickets at Initialization

AMP: Automatically Finding Model Parallel Strategies with Heterogeneity Awareness

CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing

M2OST: Many-to-one Regression for Predicting Spatial Transcriptomics from Digital Pathology Images

FedNAR: Federated Optimization with Normalized Annealing Regularization

Understanding the Forgetting of (Replay-based) Continual Learning via Feature Learning: Angle Matters

Position: TrustLLM: Trustworthiness in Large Language Models

Maestro: Uncovering Low-Rank Structures via Trainable Decomposition

Region-aware Anchoring Mechanism for Efficient Referring Visual Grounding

papers (10)

Attack of the Tails: Yes, You Really Can Backdoor Federated Learning

Rare Gems: Finding Lottery Tickets at Initialization

AMP: Automatically Finding Model Parallel Strategies with Heterogeneity Awareness

CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing

M2OST: Many-to-one Regression for Predicting Spatial Transcriptomics from Digital Pathology Images

FedNAR: Federated Optimization with Normalized Annealing Regularization

Understanding the Forgetting of (Replay-based) Continual Learning via Feature Learning: Angle Matters

Position: TrustLLM: Trustworthiness in Large Language Models

Maestro: Uncovering Low-Rank Structures via Trainable Decomposition

Region-aware Anchoring Mechanism for Efficient Referring Visual Grounding