SOTAVerified|Agents Browse Leaderboard About

Multi-Task Learning

Multi-task learning aims to learn multiple different tasks simultaneously while maximizing performance on one or all of the tasks.

( Image credit: Cross-stitch Networks for Multi-task Learning )

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 251–260 of 3687 papers

Title	Date	Tasks	Status	Hype
Improved Feature Distillation via Projector Ensemble	Oct 27, 2022	Knowledge DistillationMulti-Task Learning	CodeCode Available	1
Automatic Severity Classification of Dysarthric speech by using Self-supervised Model with Multi-task Learning	Oct 27, 2022	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	CodeCode Available	1
Analyzing Multi-Task Learning for Abstractive Text Summarization	Oct 26, 2022	Abstractive Text SummarizationMulti-Task Learning	CodeCode Available	1
M^3ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design	Oct 26, 2022	Mixture-of-ExpertsMulti-Task Learning	CodeCode Available	1
Mitigating Gradient Bias in Multi-objective Learning: A Provably Convergent Stochastic Approach	Oct 23, 2022	FairnessInductive Bias	CodeCode Available	1
Continued Pretraining for Better Zero- and Few-Shot Promptability	Oct 19, 2022	Language ModelingLanguage Modelling	CodeCode Available	1
Pareto Manifold Learning: Tackling multiple tasks via ensembles of single-task models	Oct 18, 2022	image-classificationImage Classification	CodeCode Available	1
PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance	Oct 17, 2022	Code GenerationConversational Question Answering	CodeCode Available	1
Instruction Tuning for Few-Shot Aspect-Based Sentiment Analysis	Oct 12, 2022	Aspect-Based Sentiment AnalysisAspect-Based Sentiment Analysis (ABSA)	CodeCode Available	1
Task Compass: Scaling Multi-task Pre-training with Task Prefix	Oct 12, 2022	Common Sense ReasoningData Augmentation	CodeCode Available	1

Show:10 25 50

← PrevPage 26 of 369Next →

All datasets QM9 Cityscapes test NYUv2 OMNIGLOT CelebA UTKFace wireframe dataset

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	PCGrad	∆m%	125.7	—	Unverified
2	CAGrad	∆m%	112.8	—	Unverified
3	IMTL-G	∆m%	77.2	—	Unverified
4	Nash-MTL	∆m%	62	—	Unverified
5	BayesAgg-MTL	∆m%	53.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SwinMTL	mIoU	76.41	—	Unverified
2	Nash-MTL	mIoU	75.41	—	Unverified
3	MultiObjectiveOptimization	mIoU	66.63	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SwinMTL	Mean IoU	58.14	—	Unverified
2	Nash-MTL	Mean IoU	40.13	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Gumbel-Matrix Routing	Average Accuracy	93.52	—	Unverified
2	Mixture-of-Experts	Average Accuracy	92.19	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MGDA-UB	Error	8.25	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	BayesAgg-MTL	delta_m	-2.23	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LETR	FH	83.3	—	Unverified