Our Publications

Dynamic Scoring with Enhanced Semantics for Training-Free Human-Object Interaction Detection

FedMVP: Federated Multimodal Visual Prompt Tuning for Vision-Language Models

On Large Multimodal Models as Open-World Image Classifiers

Superpowering Open-Vocabulary Object Detectors for X-ray Vision

Automatic benchmarking of large multimodal models via iterative experiment programming

Diversified in-domain synthesis with efficient fine-tuning for few-shot classification

Classifier-to-Bias: Toward Unsupervised Automatic Bias Detection for Visual Classifiers

Compositional Caching for Training-free Open-vocabulary Attribute Detection