Jing Ma

papers

352

total citations

papers (12)

VideoAutoArena: An Automated Arena for Evaluating Large Multimodal Models in Video Analysis through User Simulation

CVPR 2025arXiv

citations

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models

AAAI 2025arXiv

citations

Aligning Logits Generatively for Principled Black-Box Knowledge Distillation

CVPR 2024arXiv

citations

Meme Trojan: Backdoor Attacks Against Hateful Meme Detection via Cross-Modal Triggers

AAAI 2025arXiv

citations

Certified Causal Defense with Generalizable Robustness

AAAI 2025arXiv

citations

Decoupled Entropy Minimization

NEURIPS 2025arXiv

citations

LexLIP: Lexicon-Bottlenecked Language-Image Pre-Training for Large-Scale Image-Text Sparse Retrieval

ICCV 2023

citations

GNNs Also Deserve Editing, and They Need It More Than Once

ICML 2024

citations

Jing Ma

papers (12)

Federated Graph Classification over Non-IID Graphs

CLEAR: Generative Counterfactual Explanations on Graphs

Improved Self-Training for Test-Time Adaptation

Coarse-to-Fine Incremental Few-Shot Learning

VideoAutoArena: An Automated Arena for Evaluating Large Multimodal Models in Video Analysis through User Simulation

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models

Aligning Logits Generatively for Principled Black-Box Knowledge Distillation

Meme Trojan: Backdoor Attacks Against Hateful Meme Detection via Cross-Modal Triggers

Certified Causal Defense with Generalizable Robustness

Decoupled Entropy Minimization

LexLIP: Lexicon-Bottlenecked Language-Image Pre-Training for Large-Scale Image-Text Sparse Retrieval

GNNs Also Deserve Editing, and They Need It More Than Once

papers (12)

Federated Graph Classification over Non-IID Graphs

CLEAR: Generative Counterfactual Explanations on Graphs

Improved Self-Training for Test-Time Adaptation

Coarse-to-Fine Incremental Few-Shot Learning

VideoAutoArena: An Automated Arena for Evaluating Large Multimodal Models in Video Analysis through User Simulation

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models

Aligning Logits Generatively for Principled Black-Box Knowledge Distillation

Meme Trojan: Backdoor Attacks Against Hateful Meme Detection via Cross-Modal Triggers

Certified Causal Defense with Generalizable Robustness

Decoupled Entropy Minimization

LexLIP: Lexicon-Bottlenecked Language-Image Pre-Training for Large-Scale Image-Text Sparse Retrieval

GNNs Also Deserve Editing, and They Need It More Than Once