Metraipolis › archive › dc_ddaee97f2e8f

Leveraging Low-Level Symbolic Competences for Unsupervised Grounding in Hallucination Detection — arXiv:2609.05025

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.05025
License (as archived)
CC0-1.0
Archived
2026-10-16T22:28:59.940Z
SHA-256 of the archived text
09c8ec59c7320fa86dcf6a80b3f140b8df96c97ebba63866ced366e147d3f269
Stable URL of this page
https://metraipolis.com/documento/dc_ddaee97f2e8f

How to cite

Leveraging Low-Level Symbolic Competences for Unsupervised Grounding in Hallucination Detection — arXiv:2609.05025 — arXiv (en), https://arxiv.org/abs/2609.05025 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ddaee97f2e8f (Metraipolis archive dc_ddaee97f2e8f, sha256 09c8ec59c7320fa86dcf6a80b3f140b8df96c97ebba63866ced366e147d3f269)
BibTeX
@misc{dcddaee97f2e8f,
  title = {Leveraging Low-Level Symbolic Competences for Unsupervised Grounding in Hallucination Detection — arXiv:2609.05025},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.05025},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ddaee97f2e8f; sha256 09c8ec59c7320fa86dcf6a80b3f140b8df96c97ebba63866ced366e147d3f269},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.05025. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Hallucination-where a language model generates outputs that are factually incorrect or unsupported by the source-is a major challenge for both prompted and fine-tuned language models. Detecting hallucinations is difficult due to the opaque reasoning processes of LLMs, which often provide little insight into why a model's output may be inaccurate. In this work, we investigate whether an LLM can use an alternative, low level, symbolic competence such as SQL for unsupervised hallucination detection in some high level task. For this, we make an LLM build an SQL database from reference documents. This SQL database is then used for reasoning over the reference and the sampled response in a hallucination detection pipeline that is grounded in the database, thereby providing a neurosymbolic checkup. On RAGTruth and DiaHalu hallucination detection datasets, we find that our approach improves on direct prediction and competes with state-of-the-art hallucination detection methods, while not requiring domain-specific fine-tuning. Instead it relies on a low-level general competence already present in LLMs. This warrants further investigation of low-level LLM competences in neurosymbolic approaches.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2609.05025 — 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 2026-10-16T22:18:11.563Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).