Amazon Science
Amazon Science
ReFinED
ReFinED is an efficient and accurate entity linking (EL) system.
tubelet-transformer
This is an official implementation of TubeR: Tubelet Transformer for Video Action Detection
gluonmm
A library of transformer models for computer vision and multi-modality research
crossmodal-contrastive-learning
CrossCLR: Cross-modal Contrastive Learning For Multi-modal Video Representations, ICCV 2021
unified-ept
A Unified Efficient Pyramid Transformer for Semantic Segmentation, ICCVW 2021
azcausal
Causal Inference in Python
ContraCLM
[ACL 2023] Code for ContraCLM: Contrastive Learning For Causal Language Model
semi-vit
PyTorch implementation of Semi-supervised Vision Transformers
mix-generation
MixGen: A New Multi-Modal Data Augmentation
probconserv
Datasets and code for results presented in the ProbConserv paper