# Optimal Multi-Task Learning at Regularization Horizon for Speech Translation Task

arXiv: https://arxiv.org/abs/2509.09701 · HTML: https://mlsanity.com/paper/2509.09701/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 36 references that are in the map.
- First posted on arXiv 4 Sep 2025.

## Builds on (most cited first, 28 of 28)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [fairseq: A Fast, Extensible Toolkit for Sequence Modeling](https://mlsanity.com/paper/1904.01038/) (2019)
- [Unsupervised Data Augmentation for Consistency Training](https://mlsanity.com/paper/1904.12848/) (2019)
- [Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation](https://mlsanity.com/paper/2004.11867/) (2020)
- [R-Drop: Regularized Dropout for Neural Networks](https://mlsanity.com/paper/2106.14448/) (2021)
- [Consistency Regularization for Generative Adversarial Networks](https://mlsanity.com/paper/1910.12027/) (2019)
- [Listen and Translate: A Proof of Concept for End-to-End Speech-to-Text Translation](https://mlsanity.com/paper/1612.01744/) (2016)
- [A Simple but Tough-to-Beat Data Augmentation Approach for Natural Language Understanding and Generation](https://mlsanity.com/paper/2009.13818/) (2020)
- [End-to-End Automatic Speech Translation of Audiobooks](https://mlsanity.com/paper/1802.04200/) (2018)
- [Cross-modal Contrastive Learning for Speech Translation](https://mlsanity.com/paper/2205.02444/) (2022)
- [STEMM: Self-learning with Speech-text Manifold Mixup for Speech Translation](https://mlsanity.com/paper/2203.10426/) (2022)
- [Tied Multitask Learning for Neural Speech Translation](https://mlsanity.com/paper/1802.06655/) (2018)
- [Cascade versus Direct Speech Translation: Do the Differences Still Make a Difference?](https://mlsanity.com/paper/2106.01045/) (2021)
- [Improving Speech Translation by Understanding and Learning from the Auxiliary Text Translation Task](https://mlsanity.com/paper/2107.05782/) (2021)
- [Learning Shared Semantic Space for Speech-to-Text Translation](https://mlsanity.com/paper/2105.03095/) (2021)
- [Stacked Acoustic-and-Textual Encoding: Integrating the Pre-trained Models into Speech Translation Encoders](https://mlsanity.com/paper/2105.05752/) (2021)
- [End-to-end Speech Translation via Cross-modal Progressive Training](https://mlsanity.com/paper/2104.10380/) (2021)
- [Improving Zero-shot Translation with Language-Independent Constraints](https://mlsanity.com/paper/1906.08584/) (2019)
- [Self-Supervised Representations Improve End-to-End Speech Translation](https://mlsanity.com/paper/2006.12124/) (2020)
- [WACO: Word-Aligned Contrastive Learning for Speech Translation](https://mlsanity.com/paper/2212.09359/) (2022)
- [Improving Speech Translation by Cross-Modal Multi-Grained Contrastive Learning](https://mlsanity.com/paper/2304.10309/) (2023)
- [Modality Adaption or Regularization? A Case Study on End-to-End Speech Translation](https://mlsanity.com/paper/2306.07650/) (2023)
- [An Empirical Study of Consistency Regularization for End-to-End Speech-to-Text Translation](https://mlsanity.com/paper/2308.14482/) (2023)
- [Cross-Modal Multi-Tasking for Speech-to-Text Translation via Hard Parameter Sharing](https://mlsanity.com/paper/2309.15826/) (2023)
- [Efficient Speech Translation with Dynamic Latent Perceivers](https://mlsanity.com/paper/2210.16264/) (2022)

## Abstract (arXiv)

End-to-end speech-to-text translation typically suffers from the scarcity of paired speech-text data. One way to overcome this shortcoming is to utilize the bitext data from the Machine Translation (MT) task and perform Multi-Task Learning (MTL). In this paper, we formulate MTL from a regularization perspective and explore how sequences can be regularized within and across modalities. By thoroughly investigating the effect of consistency regularization (different modality) and R-drop (same modality), we show how they respectively contribute to the total regularization. We also demonstrate that the coefficient of MT loss serves as another source of regularization in the MTL setting. With these three sources of regularization, we introduce the optimal regularization contour in the high-dimensional space, called the regularization horizon. Experiments show that tuning the hyperparameters within the regularization horizon achieves near state-of-the-art performance on the MuST-C dataset.