Multi Task Learning

Multi-task learning (MTL) aims to improve model efficiency and generalization by training a single model to perform multiple related tasks simultaneously. Current research focuses on addressing challenges like task interference and optimization difficulties, exploring architectures such as Mixture-of-Experts (MoE), low-rank adaptors, and hierarchical models to enhance performance and efficiency across diverse tasks. MTL's significance lies in its potential to improve resource utilization and create more robust and adaptable AI systems, with applications spanning various fields including natural language processing, computer vision, and scientific modeling.

Papers

November 11, 2022

Helping the Weak Makes You Strong: Simple Multi-Task Learning Improves Non-Autoregressive Translators
Xinyou Wang, Zaixiang Zheng, Shujian Huang
Multi Task Learning Neural Machine Translation Model Non Autoregressive Translation Autoregressive Decoder

November 6, 2022

Preserving background sound in noise-robust voice conversion via multi-task learning
Jixun Yao, Yi Lei, Qing Wang, Pengcheng Guo, Ziqian Ning, Lei Xie, Hai Li, Junhui Liu, Danming Xie
Multi Task Learning Source Separation Separation Model Background Sound

November 2, 2022

Multi-task Learning for Source Attribution and Field Reconstruction for Methane Monitoring
Arka Daw, Kyongmin Yeo, Anuj Karpatne, Levente Klein
Multi Task Learning Source Attribution Field Reconstruction Dispersion Profile Sparse Measurement Methane Source Methane Emission Methane Flux

October 31, 2022

Effective Cross-Task Transfer Learning for Explainable Natural Language Inference with T5
Irina Bigoulaeva, Rachneet Sachdeva, Harish Tayyar Madabushi, Aline Villavicencio, Iryna Gurevych
Fine Tuning Multi Task Learning Natural Language Inference T5 Model Cross Task Knowledge Sequential Multi Task Learning

October 27, 2022

October 26, 2022

October 23, 2022

Mitigating Gradient Bias in Multi-objective Learning: A Provably Convergent Stochastic Approach
Heshan Fernando, Han Shen, Miao Liu, Subhajit Chaudhury, Keerthiram Murugesan, Tianyi Chen
Multi Task Learning Multi Objective Optimization Multi Objective Convergent Algorithm Biased Gradient

October 22, 2022

Extractive Summarization of Legal Decisions using Multi-task Learning and Maximal Marginal Relevance
Abhishek Agarwal, Shanshan Xu, Matthias Grabmair
Multi Task Learning Text Summarization Rhetorical Role Summarization Ability Case Summarization Maximal Marginal Relevance

October 18, 2022

Pareto Manifold Learning: Tackling multiple tasks via ensembles of single-task models
Nikolaos Dimitriadis, Pascal Frossard, François Fleuret
Multi Task Learning Diverse Ensemble Pareto Front Single Task Multi Task Benchmark Multiple Task Pareto Manifold

October 17, 2022

PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance
Yang Deng, Wenqiang Lei, Wenxuan Zhang, Wai Lam, Tat-Seng Chua
Multi Task Learning Question Answering Question Generation Table Semantics Conversational QA

October 14, 2022

Multi-Task Learning based Video Anomaly Detection with Attention
Mohammad Baradaran, Robert Bergevin
Multi Task Learning Human Attention Video Anomaly Detection Video Anomaly Motion Artifact

October 13, 2022

October 12, 2022

Multi Task Learning

Papers

Helping the Weak Makes You Strong: Simple Multi-Task Learning Improves Non-Autoregressive Translators

Preserving background sound in noise-robust voice conversion via multi-task learning

Multi-task Learning for Source Attribution and Field Reconstruction for Methane Monitoring

Effective Cross-Task Transfer Learning for Explainable Natural Language Inference with T5

Multi-task Bias-Variance Trade-off Through Functional Constraints

Automatic Severity Classification of Dysarthric speech by using Self-supervised Model with Multi-task Learning

Is Multi-Task Learning an Upper Bound for Continual Learning?

M$^3$ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design

Analyzing Multi-Task Learning for Abstractive Text Summarization

Mitigating Gradient Bias in Multi-objective Learning: A Provably Convergent Stochastic Approach

Extractive Summarization of Legal Decisions using Multi-task Learning and Maximal Marginal Relevance

Pareto Manifold Learning: Tackling multiple tasks via ensembles of single-task models

PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance

Multi-Task Learning based Video Anomaly Detection with Attention

Multi-Task Learning for Joint Semantic Role and Proto-Role Labeling

Composite Learning for Robust and Effective Dense Predictions

Entity Tracking via Effective Use of Multi-Task Learning Model and Mention-guided Decoding

Semantic Cross Attention for Few-shot Learning

Task Compass: Scaling Multi-task Pre-training with Task Prefix

Optimizing Evaluation Metrics for Multi-Task Learning via the Alternating Direction Method of Multipliers