declare-lab/MM-Align
[EMNLP 2022] This repository contains the official implementation of the paper "MM-Align: Learning Optimal Transport-based Alignment Dynamics for Fast and Accurate Inference on Missing Modality Sequences"
PythonMIT
Issues
- 3
open pre-trained weights?
#5 opened by DiTo97 - 5
why stopping at bi-modality?
#4 opened by DiTo97 - 4
- 1
word tokenizer for MELD
#3 opened by dondongwon - 0
How to align video audio with text data?
#1 opened by mrFocusXin