# AutoSeM: Automatic Task Selection and Mixing in Multi-Task Learning

arXiv: https://arxiv.org/abs/1904.04153 · HTML: https://mlsanity.com/paper/1904.04153/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Apr 2019.

## Built on by (most cited first, 22 of 22)

- [ExT5: Towards Extreme Multi-Task Scaling for Transfer Learning](https://mlsanity.com/paper/2111.10952/) (2021)
- [MURAL: Multimodal, Multitask Retrieval Across Languages](https://mlsanity.com/paper/2109.05125/) (2021)
- [Conditional Language Policy: A General Framework for Steerable Multi-Objective Finetuning](https://mlsanity.com/paper/2407.15762/) (2024)
- [A Survey of Multi-task Learning in Natural Language Processing: Regarding Task Relatedness and Training Methods](https://mlsanity.com/paper/2204.03508/) (2022)
- [Robust Multi-Objective Preference Alignment with Online DPO](https://mlsanity.com/paper/2503.00295/) (2025)
- [Identification of Negative Transfers in Multitask Learning Using Surrogate Models](https://mlsanity.com/paper/2303.14582/) (2023)
- [TinyR1-32B-Preview: Boosting Accuracy with Branch-Merge Distillation](https://mlsanity.com/paper/2503.04872/) (2025)
- [On the Complementarity of Data Selection and Fine Tuning for Domain Adaptation](https://mlsanity.com/paper/2109.07591/) (2021)
- [GradTS: A Gradient-Based Automatic Auxiliary Task Selection Method Based on Transformer Networks](https://mlsanity.com/paper/2109.05748/) (2021)
- [Evaluating Online Bandit Exploration In Large-Scale Recommender System](https://mlsanity.com/paper/2304.02572/) (2023)
- [Unsupervised Melody-to-Lyric Generation](https://mlsanity.com/paper/2305.19228/) (2023)
- [Latent-Optimized Adversarial Neural Transfer for Sarcasm Detection](https://mlsanity.com/paper/2104.09261/) (2021)
- [Towards Better Multi-task Learning: A Framework for Optimizing Dataset Combinations in Large Language Models](https://mlsanity.com/paper/2412.11455/) (2024)
- [Analyzing Multi-Task Learning for Abstractive Text Summarization](https://mlsanity.com/paper/2210.14606/) (2022)
- [Selecting Subsets of Source Data for Transfer Learning with Applications in Metal Additive Manufacturing](https://mlsanity.com/paper/2401.08715/) (2024)
- [Challenges and Opportunities of Using Transformer-Based Multi-Task Learning in NLP Through ML Lifecycle: A Survey](https://mlsanity.com/paper/2308.08234/) (2023)
- [Task Selection and Assignment for Multi-modal Multi-task Dialogue Act Classification with Non-stationary Multi-armed Bandits](https://mlsanity.com/paper/2309.09832/) (2023)
- [αVIL: Learning to Leverage Auxiliary Tasks for Multitask Learning](https://mlsanity.com/paper/2405.07769/) (2024)
- [Learning to Select the Best Forecasting Tasks for Clinical Outcome Prediction](https://mlsanity.com/paper/2407.19359/) (2024)
- [Generating Auxiliary Tasks with Reinforcement Learning](https://mlsanity.com/paper/2510.22940/) (2025)
- [BandiK: Efficient Multi-Task Decomposition Using a Multi-Bandit Framework](https://mlsanity.com/paper/2512.24708/) (2025)
- [Semi-overlapping Multi-bandit Best Arm Identification for Sequential Support Network Learning](https://mlsanity.com/paper/2512.24959/) (2025)

## Abstract (arXiv)

Multi-task learning (MTL) has achieved success over a wide range of problems, where the goal is to improve the performance of a primary task using a set of relevant auxiliary tasks. However, when the usefulness of the auxiliary tasks w.r.t. the primary task is not known a priori, the success of MTL models depends on the correct choice of these auxiliary tasks and also a balanced mixing ratio of these tasks during alternate training. These two problems could be resolved via manual intuition or hyper-parameter tuning over all combinatorial task choices, but this introduces inductive bias or is not scalable when the number of candidate auxiliary tasks is very large. To address these issues, we present AutoSeM, a two-stage MTL pipeline, where the first stage automatically selects the most useful auxiliary tasks via a Beta-Bernoulli multi-armed bandit with Thompson Sampling, and the second stage learns the training mixing ratio of these selected auxiliary tasks via a Gaussian Process based Bayesian optimization framework. We conduct several MTL experiments on the GLUE language understanding tasks, and show that our AutoSeM framework can successfully find relevant auxiliary tasks and automatically learn their mixing ratio, achieving significant performance boosts on several primary tasks. Finally, we present ablations for each stage of AutoSeM and analyze the learned auxiliary task choices.