thuml/awesome-multi-task-learning
A curated list of DATASETS, CODEBASES and PAPERS on Multi-Task Learning (MTL), from Machine Learning perspective.
Organizes MTL research across six dimensions—architectures (parameter sharing, modularity, task representation), optimization strategies (gradient balancing, task interference), datasets spanning vision/NLP/robotics/graphs, and theoretical foundations—with taxonomy covering hard/soft parameter sharing, mixture-of-experts routing, Pareto optimization, and task relationship learning. Aggregates benchmarks like Taskonomy (26 vision tasks), decaNLP, and MetaWorld alongside reference implementations in PyTorch (LibMTL, TorchJD) and domain-specific codebases for vision and recommendation systems.
822 stars.
Stars
822
Forks
65
Language
—
License
—
Category
Last pushed
Mar 03, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/thuml/awesome-multi-task-learning"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
qdrant/quaterion
Blazing fast framework for fine-tuning similarity learning models
median-research-group/LibMTL
A PyTorch Library for Multi-Task Learning
zeiss-microscopy/OAD
Collection of tools and scripts useful to automate microscopy workflows in ZEN Blue using Python...
lucidrains/PoPE-pytorch
Efficient implementation (and explorations) into polar coordinate positional embedding (PoPE) -...
Zehong-Wang/Awesome-Weight-Space-Learning
A collection of weight space learning including papers, codes, and datasets.