NAVER LABS System Re-implementation for the IWSLT 2026 Instruction-Following Task

Anand Kamble; Aniket Tathe

NAVER LABS System Re-implementation for the IWSLT 2026 Instruction-Following Task

Computation and Language 2026-07-06 v1

Authors: Anand Kamble , Aniket Tathe

Abstract

We re-implement the NAVER LABS IWSLT 2025 instruction-following pipeline for the IWSLT 2026 Shared Task (constrained condition, short audio track), adapting it to the mandated components: SeamlessM4T-v2-large as the speech encoder and Qwen3-4B-Instruct as the LLM backbone. The three-stage approach projector alignment, text-only LoRA pre-training, and multimodal merging is preserved from the original design. We additionally construct 100k synthetic instruction-following examples across ten speech-centric task types (10k per task) from the provided corpora, suitable for further Stage 3 fine-tuning. Our primary model achieves COMET 0.781 on EN-ZH speech translation and BERTScore-F1 0.346 on English SQA on the MCIF benchmark.

Keywords

speech translation instruction following multimodal large language models LoRA speech encoder shared task

Cite

@article{arxiv.2607.05623,
  title  = {NAVER LABS System Re-implementation for the IWSLT 2026 Instruction-Following Task},
  author = {Anand Kamble and Aniket Tathe},
  journal= {arXiv preprint arXiv:2607.05623},
  year   = {2026}
}