# Monotonic Multihead Attention

arXiv: https://arxiv.org/abs/1909.12406 · HTML: https://mlsanity.com/paper/1909.12406/

## Key facts

- Cited by 93 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Sep 2019.

## Built on by (most cited first, 30 of 93)

- [LLaMA-Omni: Seamless Speech Interaction with Large Language Models](https://mlsanity.com/paper/2409.06666/) (2024)
- [Translating Images into Maps](https://mlsanity.com/paper/2110.00966/) (2021)
- [Information-Transport-based Policy for Simultaneous Translation](https://mlsanity.com/paper/2210.12357/) (2022)
- [TransLLaMa: LLM-based Simultaneous Translation System](https://mlsanity.com/paper/2402.04636/) (2024)
- [Universal Simultaneous Machine Translation with Mixture-of-Experts Wait-k Policy](https://mlsanity.com/paper/2109.05238/) (2021)
- [Attention as a Guide for Simultaneous Speech Translation](https://mlsanity.com/paper/2212.07850/) (2022)
- [Large-Scale Streaming End-to-End Speech Translation with Neural Transducers](https://mlsanity.com/paper/2204.05352/) (2022)
- [Learning When to Translate for Streaming Speech](https://mlsanity.com/paper/2109.07368/) (2021)
- [Mask Attention Networks: Rethinking and Strengthen Transformer](https://mlsanity.com/paper/2103.13597/) (2021)
- [SimulSLT: End-to-End Simultaneous Sign Language Translation](https://mlsanity.com/paper/2112.04228/) (2021)
- [Hidden Markov Transformer for Simultaneous Machine Translation](https://mlsanity.com/paper/2303.00257/) (2023)
- [Does Simultaneous Speech Translation need Simultaneous Models?](https://mlsanity.com/paper/2204.03783/) (2022)
- [End-to-End Simultaneous Speech Translation with Differentiable Segmentation](https://mlsanity.com/paper/2305.16093/) (2023)
- [Modeling Dual Read/Write Paths for Simultaneous Machine Translation](https://mlsanity.com/paper/2203.09163/) (2022)
- [StreamSpeech: Simultaneous Speech-to-Speech Translation with Multi-task Learning](https://mlsanity.com/paper/2406.03049/) (2024)
- [Hybrid Transducer and Attention based Encoder-Decoder Modeling for Speech-to-Text Tasks](https://mlsanity.com/paper/2305.03101/) (2023)
- [Gaussian Multi-head Attention for Simultaneous Machine Translation](https://mlsanity.com/paper/2203.09072/) (2022)
- [Reducing Position Bias in Simultaneous Machine Translation with Length-Aware Framework](https://mlsanity.com/paper/2203.09053/) (2022)
- [Wait-info Policy: Balancing Source and Target at Information Level for Simultaneous Machine Translation](https://mlsanity.com/paper/2210.11220/) (2022)
- [Learning Optimal Policy for Simultaneous Machine Translation via Binary Search](https://mlsanity.com/paper/2305.12774/) (2023)
- [An Empirical Study of End-to-end Simultaneous Speech Translation Decoding Strategies](https://mlsanity.com/paper/2103.03233/) (2021)
- [Exploring Continuous Integrate-and-Fire for Adaptive Simultaneous Speech Translation](https://mlsanity.com/paper/2204.09595/) (2022)
- [Turning Fixed to Adaptive: Integrating Post-Evaluation into Simultaneous Machine Translation](https://mlsanity.com/paper/2210.11900/) (2022)
- [Chronological Thinking in Full-Duplex Spoken Dialogue Language Models](https://mlsanity.com/paper/2510.05150/) (2025)
- [The USTC-NELSLIP Systems for Simultaneous Speech Translation Task at IWSLT 2021](https://mlsanity.com/paper/2107.00279/) (2021)
- [SiLLM: Large Language Models for Simultaneous Machine Translation](https://mlsanity.com/paper/2402.13036/) (2024)
- [A Weakly-Supervised Streaming Multilingual Speech Model with Truly Zero-Shot Capability](https://mlsanity.com/paper/2211.02499/) (2022)
- [Non-autoregressive Streaming Transformer for Simultaneous Translation](https://mlsanity.com/paper/2310.14883/) (2023)
- [Blockwise Streaming Transformer for Spoken Language Understanding and Simultaneous Speech Translation](https://mlsanity.com/paper/2204.08920/) (2022)
- [Decoder-only Streaming Transformer for Simultaneous Translation](https://mlsanity.com/paper/2406.03878/) (2024)

## Abstract (arXiv)

Simultaneous machine translation models start generating a target sequence before they have encoded or read the source sequence. Recent approaches for this task either apply a fixed policy on a state-of-the art Transformer model, or a learnable monotonic attention on a weaker recurrent neural network-based structure. In this paper, we propose a new attention mechanism, Monotonic Multihead Attention (MMA), which extends the monotonic attention mechanism to multihead attention. We also introduce two novel and interpretable approaches for latency control that are specifically designed for multiple attentions heads. We apply MMA to the simultaneous machine translation task and demonstrate better latency-quality tradeoffs compared to MILk, the previous state-of-the-art approach. We also analyze how the latency controls affect the attention span and we motivate the introduction of our model by analyzing the effect of the number of decoder layers and heads on quality and latency.