AlignAtt4LLM: Fast AlignAtt for Decoder-Only LLMs at IWSLT 2026 Simultaneous Speech Translation Task
Not on the map. Why?
Abstract (arXiv)
We describe AlignAtt4LLM, an IWSLT 2026 simultaneous speech translation system for English to German, Italian, and Chinese. The system is a synchronous cascade: Qwen3-ASR with forced alignment produces an incrementally updated source transcript, and Gemma-4 E4B-it translates that prefix under an MT-side AlignAtt policy. To our knowledge, this is the first application of AlignAtt to a decoder-only LLM, where the encoder-decoder cross-attention used by earlier AlignAtt systems is absent. We recover a usable policy by proposing (1) an explicit source span in the prompt, (2) offline selection of translation-specific alignment heads, (3) selective qk-fast replay of the draft-to-source attention block, and (4) runtime query/key capture that preserves model outputs bit-identically. On the IWSLT 2026 development set, AlignAtt4LLM outperforms the supplied baselines for the European target languages, English to German and English to Italian, in both the low-latency regime around 2 seconds and the high-latency regime below 4 seconds CU-LongYAAL. Results for English to Chinese are more mixed, but the method is not tied to Gemma-4: because AlignAtt4LLM only requires a deterministic prompt layout, calibrated attention heads, and query/key capture, the same policy can be reapplied to stronger translation-focused decoder-only MT backbones for non-European target languages.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 11 of its 19 references that are in the map. First posted on arXiv 2 Jun 2026.
Builds on
11 of 19 references.
References · 11
- 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
- 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
- 2023xCOMET: Transparent Machine Translation Evaluation through Fine-grained Error Detectioncited by 265
- 2023AlignAtt: Using Attention-based Audio-Translation Alignments as a Guide for Simultaneous Speech Translationcited by 35
- 2022CUNI-KIT System for Simultaneous Speech Translation Task at IWSLT 2022cited by 30
- 2024StreamAtt: Direct Streaming Speech-to-Text Translation with Attention-based Audio History Selectioncited by 19
- 2023Incremental Blockwise Beam Search for Simultaneous Speech Translation with Controllable Quality-Latency Tradeoffcited by 8
- 2025Simultaneous Translation with Offline Speech and LLM Models in CUNI Submission to IWSLT 2025cited by 8
- 2025Better Late Than Never: Meta-Evaluation of Latency Metrics for Simultaneous Speech-to-Text Translationcited by 4
- 2025Simulstream: Open-Source Toolkit for Evaluation and Demonstration of Streaming Speech-to-Text Translation Systemscited by 4
- 2026Scaling Model and Data for Multilingual Machine Translation with Open Large Language Modelscited by 4
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025PART: Progressive Alignment Representation Training for Multilingual Speech-To-Text with LLMs
- 2025Instituto de Telecomunicacc~oes at IWSLT 2025: Aligning Small-Scale Speech and Language Models for Speech-to-Text Learning
- 2025Adaptive Inner Speech-Text Alignment for LLM-based Speech Translation
- 2025POTSA: A Cross-Lingual Speech Alignment Framework for Speech-to-Text Translation
- 2024Decoding-Time Language Model Alignment with Multiple Objectives
- 2025A Survey on Training-free Alignment of Large Language Models
- 2025LatinX: Aligning a Multilingual TTS Model with Direct Preference Optimization
- 2024Tuning LLMs with Contrastive Alignment Instructions for Machine Translation in Unseen, Low-resource Languages