# DOA: Training-Free Decoder-Only Attention Policy for Long-Form Simultaneous Translation with SpeechLLMs

arXiv: https://arxiv.org/abs/2605.31432 · HTML: https://mlsanity.com/paper/2605.31432/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 35 references that are in the map.
- First posted on arXiv 29 May 2026.

## Builds on (most cited first, 23 of 23)

- [Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs](https://mlsanity.com/paper/2503.01743/) (2025)
- [On decoder-only architecture for speech-to-text and large language model integration](https://mlsanity.com/paper/2307.03917/) (2023)
- [SimulMT to SimulST: Adapting Simultaneous Text Translation to End-to-End Simultaneous Speech Translation](https://mlsanity.com/paper/2011.02048/) (2020)
- [AlignAtt: Using Attention-based Audio-Translation Alignments as a Guide for Simultaneous Speech Translation](https://mlsanity.com/paper/2305.11408/) (2023)
- [Attention as a Guide for Simultaneous Speech Translation](https://mlsanity.com/paper/2212.07850/) (2022)
- [LLM-Inference-Bench: Inference Benchmarking of Large Language Models on AI Accelerators](https://mlsanity.com/paper/2411.00136/) (2024)
- [StreamAtt: Direct Streaming Speech-to-Text Translation with Attention-based Audio History Selection](https://mlsanity.com/paper/2406.06097/) (2024)
- [BESTOW: Efficient and Streamable Speech Language Model with the Best of Two Worlds in GPT and T5](https://mlsanity.com/paper/2406.19954/) (2024)
- [Findings of the IWSLT 2024 Evaluation Campaign](https://mlsanity.com/paper/2411.05088/) (2024)
- [How "Real" is Your Real-Time Simultaneous Speech-to-Text Translation System?](https://mlsanity.com/paper/2412.18495/) (2024)
- [MCIF: Multimodal Crosslingual Instruction-Following Benchmark from Scientific Talks](https://mlsanity.com/paper/2507.19634/) (2025)
- [From Simultaneous to Streaming Machine Translation by Leveraging Streaming History](https://mlsanity.com/paper/2203.02459/) (2022)
- [Investigating Decoder-only Large Language Models for Speech-to-text Translation](https://mlsanity.com/paper/2407.03169/) (2024)
- [Simul-Whisper: Attention-Guided Streaming Whisper with Truncation Detection](https://mlsanity.com/paper/2406.10052/) (2024)
- [FASST: Fast LLM-based Simultaneous Speech Translation](https://mlsanity.com/paper/2408.09430/) (2024)
- [Hearing to Translate: The Effectiveness of Speech Modality Integration into LLMs](https://mlsanity.com/paper/2512.16378/) (2025)
- [Long-Form End-to-End Speech Translation via Latent Alignment Segmentation](https://mlsanity.com/paper/2309.11384/) (2023)
- [StreamUni: Achieving Streaming Speech Translation with a Unified Large Speech-Language Model](https://mlsanity.com/paper/2507.07803/) (2025)
- [What Helps Transformers Recognize Conversational Structure? Importance of Context, Punctuation, and Labels in Dialog Act Recognition](https://mlsanity.com/paper/2107.02294/) (2021)
- [Exploring the limits of decoder-only models trained on public speech recognition corpora](https://mlsanity.com/paper/2402.00235/) (2024)
- [Large Language Models Are Read/Write Policy-Makers for Simultaneous Generation](https://mlsanity.com/paper/2501.00868/) (2025)
- [Simulstream: Open-Source Toolkit for Evaluation and Demonstration of Streaming Speech-to-Text Translation Systems](https://mlsanity.com/paper/2512.17648/) (2025)
- [Semantic Segmentation with Bidirectional Language Models Improves Long-form ASR](https://mlsanity.com/paper/2305.18419/) (2023)

## Built on by (most cited first, 1 of 1)

- [Gradient-Based Speech-to-Text Alignment for Any ASR Model: From CTC to Speech LLMs](https://mlsanity.com/paper/2607.06831/) (2026)

## Abstract (arXiv)

Simultaneous speech-to-text translation (SimulST) generates translations while speech is still unfolding, requiring a streaming policy that decides when to read and when to write. State-of-the-art approaches rely on attention-based encoder-decoder models where cross-attention provides explicit alignment signals. In contrast, Speech Large Language Models (SpeechLLMs) are decoder-only architectures relying solely on self-attention. This raises a central question: whether decoder self-attention contains sufficiently stable alignment signals to guide the streaming policy. Moreover, existing approaches typically rely on training-based adaptations or heuristic wait-k policies and have not been validated in long-form settings. To fill these gaps, we propose Decoder-Only Attention (DOA), a training-free policy that enables long-form simultaneous translation with off-the-shelf SpeechLLMs by deriving a proxy alignment from self-attention. Experiments on Phi4-Multimodal and Qwen3-Omni show that DOA provides an effective alignment signal for supporting streaming decisions, enabling low-latency long-form SimulST with quality close to offline decoding without retraining.