Metraipolis › archive › dc_dce7f9459bdd

Leading-Silence Augmentation and Multi-Stage Synthetic Supervision for the Second MLC-SLM Challenge — arXiv:2608.14150

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.14150
License (as archived)
CC0-1.0
Archived
2026-10-19T02:27:56.150Z
SHA-256 of the archived text
1f8506b7b2055a233df4f8fb92a24f270b1e260679c737423a5da02ff05c5c6f
Stable URL of this page
https://metraipolis.com/documento/dc_dce7f9459bdd

How to cite

Leading-Silence Augmentation and Multi-Stage Synthetic Supervision for the Second MLC-SLM Challenge — arXiv:2608.14150 — arXiv (en), https://arxiv.org/abs/2608.14150 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dce7f9459bdd (Metraipolis archive dc_dce7f9459bdd, sha256 1f8506b7b2055a233df4f8fb92a24f270b1e260679c737423a5da02ff05c5c6f)
BibTeX
@misc{dcdce7f9459bdd,
  title = {Leading-Silence Augmentation and Multi-Stage Synthetic Supervision for the Second MLC-SLM Challenge — arXiv:2608.14150},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14150},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dce7f9459bdd; sha256 1f8506b7b2055a233df4f8fb92a24f270b1e260679c737423a5da02ff05c5c6f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14150. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The second Multilingual Conversational Speech Language Model (MLC-SLM) Challenge evaluates two tasks over complete, unsegmented multilingual conversations: speaker diarization and recognition (Task 1) and conversational speech understanding (Task 2). Neither task provides oracle utterance boundaries or speaker labels at evaluation, and Task 2 provides no question-answer training set. For Task 1, we fine-tune VibeVoice-ASR-7B with random leading-silence cropping, consistent timestamp correction, and an exponential moving average (EMA) training strategy. For Task 2, we construct synthetic question-answer pairs through multimodal candidate generation, silent-audio filtering, and distribution-matched augmentation, and fine-tune Qwen3-Omni-30B-A3B-Instruct for tagged direct answering. On the Task 1 evaluation set, cropping reduces tcpMER from 18.30% to 17.27%, and EMA further reduces it to 16.73%. On the Task 2 evaluation set, jointly applying distribution-matched augmentation and tagged direct answering raises accuracy from 83.0% to 86.0%.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.14150 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T02:20:31.907Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).