mlsanityData to 14 Oct 2026

FBK's Long-form SpeechLLMs for IWSLT 2026 Instruction Following

arXiv 2606.26819 ↗First posted 25 Jun 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper describes our submission to the IWSLT 2026 Instruction Following shared task. SpeechLLMs are developed for both short-form and long-form speech instruction following under constrained settings. For the short track, strong performance is achieved on MCIF, with a SIFS score of 2.0708. For the long track, three speech segmentation methods are explored, and the HIFS score is introduced to account for unstable long-form generation. Experimental results show that fixed 30-second segmentation provides the most robust long-form performance, achieving the highest HIFS score of 2.0663. Further analysis shows that hallucination mainly manifests as repetitive insertions in generated outputs, substantially affecting ASR and SSUM, while short-form capabilities are largely retained after long-form extension.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 21 references that are in the map. First posted on arXiv 25 Jun 2026.

Builds on

15 of 21 references.

References · 15

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2019Transformer-XL: Attentive Language Models Beyond a Fixed-Length Contextcited by 1,850
  3. 2025Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAscited by 521
  4. 2021SpeechBrain: A General-Purpose Speech Toolkitcited by 406
  5. 2019Europarl-ST: A Multilingual Corpus For Speech Translation Of Parliamentary Debatescited by 114
  6. 2023AudioChatLlama: Towards General-Purpose Speech Abilities for LLMscited by 59
  7. 2022GigaST: A 10,000-hour Pseudo Speech Translation Corpuscited by 20
  8. 2021Beyond Voice Activity Detection: Hybrid Audio Segmentation for Direct Speech Translationcited by 17
  9. 2024From Text Segmentation to Smart Chaptering: A Novel Benchmark for Structuring Video Transcriptionscited by 16
  10. 2023Investigating End-to-End ASR Architectures for Long Form Audio Transcriptioncited by 11
  11. 2023Updated Corpora and Benchmarks for Long-Form Speech Recognitioncited by 10
  12. 2024Investigating Decoder-only Large Language Models for Speech-to-text Translationcited by 9
3 more
  1. 2025Prepending or Cross-Attention for Speech-to-Text? An Empirical Comparisoncited by 5
  2. 2025NAVER LABS Europe Submission to the Instruction-following Trackcited by 5
  3. 2025KIT's Offline Speech Translation and Instruction Following Submission for IWSLT 2025cited by 4

Built on by

Not cited yet.