SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.10772 · ICCV · 2021

UniT: Multimodal Multitask Learning with a Unified Transformer

Ronghang Hu

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

VE detection QNLI SST-2 VQA MNLI SNLI-VE QNLI VQA detection QQP SST-2 QQP MNLI SNLI-VE detection QNLI SST-2 VQA MNLI SNLI-VE QNLI VQA detection QQP SST-2 Figure 1: In this work, we propose UniT, which jointly learns multiple tasks across different domains with a Unified Transformer. Our UniT model simultaneously handles 7 tasks on 8 datasets ranging from object detection to vision-andlanguage reasoning and natural language understanding, while achieving strong performance on each task with a compact set of model parameters.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2102.10772")
get_code_for_paper("2102.10772")
have("2102.10772")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2102.10772.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2102.10772.svg)](https://syntology.ai/paper/2102.10772)

Connect an agent — have() is free.