Trending

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and more.Subscribe

Added to collectionTop 100 Deep Learning and AI Papers 2021

Computer Vision

Computation and Language

UniT: A Unified Transformer Model for Multi-Task Learning

Transformer is All You Need: Multimodal Multitask Learning with a Unified Transformer

We propose a unified transformer model to simultaneously learn the most prominent tasks across different domains, ranging from object detection to language understanding and multimodal reasoning.

Our model encodes each input modality with an encoder and makes predictions on each task with a shared decoder over the encoded input representations, followed by task-specific output heads.

The entire model is jointly trained end-to-end with losses from each task.

In our experiments, we learn 7 tasks jointly over 8 datasets, achieving comparable performance to well-established prior work on each domain under the same supervision with a compact set of model parameters.

Authors

Ronghang Hu, Amanpreet Singh

Related Topics

Transformer encoder-decoder architecture

Task-specific models

Unified transformer model

Read the Paper

Read Paper

Trending

Bookmarks

◐ Recommended Follows

Julien Chaumond

More

◐ Latest Activity

More

UniT: A Unified Transformer Model for Multi-Task Learning - 42Papers

Read Related Papers

Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes

A Literature Review of Transfer, Meta, and Multi-Task Learning

Temporal Modeling via Video Transformer and Masked Visual-token Modeling

Utilizing Resource-Rich Language Datasets for End-to-End Scene Text Recognition in Resource-Poor Languages

Learning Event-Based Models via Knowledge Distillation

PeCo: Perceptual Codebook for BERT Pre-training of Vision Transformers

Cerberus: An Attention-Based Architecture for Multi-Task Indoor Scene Understanding

NÜWA: Visual Synthesis Pre-training for Neural visUal World creAtion

Scaling Up Vision-Language Pre-training for Image Captioning

Self-slimmed Vision Transformer