北京通用人工智能研究院BIGAI

科研成果

Latent Diffusion Energy-Based Model for Interpretable Text Modeling

Weakly Supervised Video Moment Localization with Contrastive Negative Sample Mining

Weakly Supervised Temporal Sentence Grounding with Gaussian-based Contrastive Proposal Learning

Spatio-temporal Self-Supervised Representation Learning for 3D Point Clouds

VLGrammar: Grounded Grammar Induction of Vision and Language

YouRefIt: Embodied Reference Understanding with Language and Gesture