HQD-EM : Robust VQA through Hierarchical Question Decomposition and Ensemble-Adaptive Margins

Preserving Multi-Modal Capabilities of Pre-trained VLMs for Improving Vision-Linguistic Compositionality

Let Me Finish My Sentence: Video Temporal Grounding with Holistic Text Understanding

Grained Action Understanding with Tools in Instructional Videos

Empirical study on using Adapters for debiased Visual Question Answering

Counterfactual Mix-Up for Visual Question Answering

Technical Report of NICE Challenge at CVPR 2023: Retrieval-based Data Discovery and Fusion for Zero-shot Image Captioning

Generative Bias for Robust Visual Question Answering

Self-Sufficient Framework for Continuous Sign Language Recognition

Signing Outside the Studio: Benchmarking Background Robustness for Continuous Sign Language Recognition