Metraipolis › archive › dc_13577991990d

Hierarchy-Aware Supervised Uncertainty Estimation for Black-box LLM Taxonomic Reasoning — arXiv:2608.22839

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.22839
License (as archived)
CC0-1.0
Archived
2026-10-18T07:45:04.637Z
SHA-256 of the archived text
187cf4b43dac0e3c1c5fc929f7d267441582d27b2fb6506e5bfa39ab93e1fd20
Stable URL of this page
https://metraipolis.com/documento/dc_13577991990d

How to cite

Hierarchy-Aware Supervised Uncertainty Estimation for Black-box LLM Taxonomic Reasoning — arXiv:2608.22839 — arXiv (en), https://arxiv.org/abs/2608.22839 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_13577991990d (Metraipolis archive dc_13577991990d, sha256 187cf4b43dac0e3c1c5fc929f7d267441582d27b2fb6506e5bfa39ab93e1fd20)
BibTeX
@misc{dc13577991990d,
  title = {Hierarchy-Aware Supervised Uncertainty Estimation for Black-box LLM Taxonomic Reasoning — arXiv:2608.22839},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.22839},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_13577991990d; sha256 187cf4b43dac0e3c1c5fc929f7d267441582d27b2fb6506e5bfa39ab93e1fd20},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.22839. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) are increasingly used for scientific decision support, yet reliable confidence estimation remains difficult in black-box settings. We study uncertainty estimation for hierarchical taxonomic reasoning generated by a black-box LLM in a long-tailed biodiversity monitoring pipeline. Using proxy features extracted by an open-source tool LLM, we train lightweight supervised estimators with hierarchy-aware supervision to predict rank-wise correctness. Across three tool LLMs, the supervised estimators consistently outperform a token-likelihood baseline for micro discrimination and selective prediction under a single global rejection threshold, improving micro AUROC from 0.57 to 0.75--0.80. The best results are achieved by a rank-specific multi-head design (H3), suggesting that accounting for hierarchical output structure is important when a unified abstention rule is required. Our code is publicly available at this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.22839 — 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 2026-10-18T07:37:52.903Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).