code for our CVPR2020 paper "IMRAM: Iterative Matching with Recurrent Attention Memory for Cross-Modal Image-Text Retrieval"

This repo contains the code for our paper "EditNTS: An Neural Programmer-Interpreter Model for Sentence Simplification through Explicit Editing" accepted at ACL 2019.

Code for CVPR 19 Paper "Improving Referring Expression Grounding with Cross-modal Attention-guided Erasing"

MAttNet: Modular Attention Network for Referring Expression Comprehension

Learning phrase grounding from captioned images through InfoNCE bound on mutual information

Multi Task Vision and Language

PyTorch source code for "Stacked Cross Attention for Image-Text Matching"

An PyTorch reimplementation of bottom-up-attention models

Code release for Learning to Assemble Neural Module Tree Networks for Visual Grounding (ICCV 2019)

Referring Expression Datasets API

Show, Control and Tell: A Framework for Generating Controllable and Grounded Captions. CVPR 2019