Metraipolis › archive › dc_fe2a48d7c8cd

A Probe Shift Is Not a Fairness Fix: The Limits of Representation Steering in Speech Models — arXiv:2609.18533

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.18533
License (as archived): CC0-1.0
Archived: 2026-10-15T13:49:29.980Z
SHA-256 of the archived text: ab066e790761267996bd5cb49dce54e99fd6fecbe1f2953d1e8733473d39d387
Stable URL of this page: https://metraipolis.com/documento/dc_fe2a48d7c8cd

How to cite

A Probe Shift Is Not a Fairness Fix: The Limits of Representation Steering in Speech Models — arXiv:2609.18533 — arXiv (en), https://arxiv.org/abs/2609.18533 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fe2a48d7c8cd (Metraipolis archive dc_fe2a48d7c8cd, sha256 ab066e790761267996bd5cb49dce54e99fd6fecbe1f2953d1e8733473d39d387)

BibTeX

@misc{dcfe2a48d7c8cd,
  title = {A Probe Shift Is Not a Fairness Fix: The Limits of Representation Steering in Speech Models — arXiv:2609.18533},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18533},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fe2a48d7c8cd; sha256 ab066e790761267996bd5cb49dce54e99fd6fecbe1f2953d1e8733473d39d387},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18533. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Automatic speech recognition (ASR) systems exhibit unequal error rates across speaker groups, motivating interventions on their internal representations. We ask whether speaker-linked attributes that are linearly readable from pretrained ASR encoders yield useful directions for reducing group word-error-rate (WER) gaps. Across Whisper-medium, HuBERT-large, and Wav2Vec2-large on Common Voice and the Speech Accent Archive, we probe every encoder layer for metadata-derived sex/gender, age, and native/accent labels; construct centroid and probe-derived directions; inject them at selected layers; and compare downstream probe trajectories with matched WER changes. Sex labels are highly decodable (best macro-F1 0.924--0.941), native/accent labels are also above chance (0.544--0.696), and age is weaker (0.354--0.397). Of 22 post-selected reruns, nine have 95% paired-bootstrap intervals entirely below zero, yet every absolute source-group WER reduction is below 0.7 percentage points. Conversely, a local target-class probe rate can rise from 8.09% to 99.87% while WER worsens. Linear readability is therefore neither evidence of causal use nor a reliable mitigation method. Our results motivate evaluating speech-bias interventions jointly at representation, propagation, and task levels.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.18533 — 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 2026-10-15T13:48:24.327Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).