Metraipolis › archive › dc_f7d2ea42bd4e

CLASH: Counterfactual Auditing of Lexical and Prosodic Reliance in Spoken Sarcasm Detection — arXiv:2609.16582

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.16582
License (as archived): CC0-1.0
Archived: 2026-10-15T19:23:27.019Z
SHA-256 of the archived text: be6b1091742b6d33a286e3d77f4824ec5419e7a35389261eef4f5ee23c25eab0
Stable URL of this page: https://metraipolis.com/documento/dc_f7d2ea42bd4e

How to cite

CLASH: Counterfactual Auditing of Lexical and Prosodic Reliance in Spoken Sarcasm Detection — arXiv:2609.16582 — arXiv (en), https://arxiv.org/abs/2609.16582 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f7d2ea42bd4e (Metraipolis archive dc_f7d2ea42bd4e, sha256 be6b1091742b6d33a286e3d77f4824ec5419e7a35389261eef4f5ee23c25eab0)

BibTeX

@misc{dcf7d2ea42bd4e,
  title = {CLASH: Counterfactual Auditing of Lexical and Prosodic Reliance in Spoken Sarcasm Detection — arXiv:2609.16582},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.16582},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f7d2ea42bd4e; sha256 be6b1091742b6d33a286e3d77f4824ec5419e7a35389261eef4f5ee23c25eab0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.16582. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Spoken sarcasm detectors may exploit lexical content, prosody, or their interaction, yet conventional evaluation cannot reveal which cues drive their predictions. We introduce CLASH (Controlled Lexical-Acoustic Separation Harness), a bilingual counterfactual diagnostic framework that evaluates each utterance under original, lexical-preserving, prosody-preserving, and approximately neutralised conditions. We evaluate handcrafted acoustic-feature systems, self-supervised learning (SSL) probes, and large audio language models (LALMs) on CMMA and MUStARD. For target-only Qwen3-Omni, lexical-preserving speech retains a 0.135--0.148 AUROC advantage over prosody-preserving speech after duration balancing, with cluster-bootstrap intervals above zero; alternative lexical resynthesis preserves this advantage. Acoustic interventions shift scores without consistently improving discrimination or changing binary predictions under the evaluated conditions. Context and interaction estimates vary across corpora. These findings distinguish acoustic sensitivity from sarcasm discrimination while exposing duration, identity, and transformation effects.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-16 in cs.CL).
Categorie arXiv: Sound (cs.SD) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.16582 — 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 2026-10-15T18:32:32.902Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).