{
  "id": 3425,
  "url": "https://arxiv.org/abs/2605.31432v1",
  "title": "DOA: Training-Free Decoder-Only Attention Policy for Long-Form Simultaneous Translation with SpeechLLMs",
  "summary": "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 con",
  "authors": "Sara Papi, Luisa Bentivogli",
  "category": "research",
  "topics": "regulation,safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-05-29T15:27:26.000Z",
  "fetched_at": "2026-07-14T16:30:14.370Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/3425",
  "original_url": "https://arxiv.org/abs/2605.31432v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}