Metraipolis › archive › dc_5b99cbf004e3
When Do Language-Grounded Explanations Help? A Graph-Bottleneck for Farm Monitoring Interpretable Sheep Facial Pain — arXiv:2609.20427
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.20427
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T03:24:17.953Z
- SHA-256 of the archived text
95ae1d28ff03e570676dee180c4b29a911dca5964baf0b40e112af1d0d0be1f8- Stable URL of this page
https://metraipolis.com/documento/dc_5b99cbf004e3
How to cite
When Do Language-Grounded Explanations Help? A Graph-Bottleneck for Farm Monitoring Interpretable Sheep Facial Pain — arXiv:2609.20427 — arXiv (en), https://arxiv.org/abs/2609.20427 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5b99cbf004e3 (Metraipolis archive dc_5b99cbf004e3, sha256 95ae1d28ff03e570676dee180c4b29a911dca5964baf0b40e112af1d0d0be1f8)
BibTeX
@misc{dc5b99cbf004e3,
title = {When Do Language-Grounded Explanations Help? A Graph-Bottleneck for Farm Monitoring Interpretable Sheep Facial Pain — arXiv:2609.20427},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.20427},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5b99cbf004e3; sha256 95ae1d28ff03e570676dee180c4b29a911dca5964baf0b40e112af1d0d0be1f8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20427. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Automated pain recognition from facial expression could make continuous welfare assessment practical in sheep, but adoption depends on trust: a stockperson cannot act on a score that arrives without justification. We ground a model in the Sheep Pain Facial Expression Scale (SPFES) by letting each detected facial region attend over text embeddings of the clinical descriptors and then test whether the resulting explanations mean anything. They do not. Ablating an entire descriptor changes the predicted logit by about $10^{-4}$, and the most-attended cue agrees with the predicted pain level in only $32.6\%$ of regions, although the attention maps, the learned gate, and the generated text all proposed otherwise. We therefore remove the appearance bypass with a concept bottleneck whose classifier reads only SPFES concept scores, supervised by per-region state annotations that image-level pipelines discard. This costs $0.05$--$0.10$ in Cohen's $\kappa$ but yields concepts that are demonstrably learned: minority pain-indicating states are recovered at $3.5$--$8.3\times$ their base rates, and the ear and eye severity orderings emerge without severity supervision. Removing the supervision alone leaves $\kappa$ unchanged while concept accuracy falls to $0.109$, showing that architectural necessity does not imply semantic validity. We also show that pooled concept accuracy is misleading under clinical imbalance and provide a cross-validated, protocol-matched benchmark of seven methods on this dataset.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.20427 — 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 2026-10-10T03:04:13.955Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).