Metraipolis › archive › dc_bced008449a3

Do Open Language Models Invent Scripture? BILAL-BENCH: Benchmarking Islamic-Source Fabrication in Bengali — arXiv:2610.40996

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.40996
License (as archived)
CC0-1.0
Archived
2026-11-03T07:34:29.652Z
SHA-256 of the archived text
9ca34834efff616084ec3bc8060dd49988bf81abac3005b1ee80629a4e83f92d
Stable URL of this page
https://metraipolis.com/documento/dc_bced008449a3

How to cite

Do Open Language Models Invent Scripture? BILAL-BENCH: Benchmarking Islamic-Source Fabrication in Bengali — arXiv:2610.40996 — arXiv (en), https://arxiv.org/abs/2610.40996 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bced008449a3 (Metraipolis archive dc_bced008449a3, sha256 9ca34834efff616084ec3bc8060dd49988bf81abac3005b1ee80629a4e83f92d)
BibTeX
@misc{dcbced008449a3,
  title = {Do Open Language Models Invent Scripture? BILAL-BENCH: Benchmarking Islamic-Source Fabrication in Bengali — arXiv:2610.40996},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.40996},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bced008449a3; sha256 9ca34834efff616084ec3bc8060dd49988bf81abac3005b1ee80629a4e83f92d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40996. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

When a language model invents a Qur'anic verse or hadith, it gives false authority to text that readers may treat as sacred. Existing Islamic hallucination benchmarks are largely Arabic and often closed-ended, while Bengali resources mainly study general-domain hallucination or factual knowledge. We introduce BILAL-BENCH, a 240-question expert-verified benchmark for open-ended Islamic-source fabrication in Bengali. It covers Qur'anic attribution, hadith, fiqh, contemporary issues, and adversarial false-premise prompts, and uses a two-axis rubric that separates source errors from assertiveness. We evaluate seven open-weight models from 1B to 120B on 1,680 double-annotated responses. Six of the seven models confidently fabricate on 79.3-95.3% of source-production questions, with no monotonic improvement from scale. Hadith provenance is the weakest category across models: only two of 315 responses are correct and correctly sourced. The two 8B BanglaLLaMA variants have similar confident-fabrication rates but different error profiles. Models vary far more in whether they withhold a claim they cannot support than in source accuracy: appropriate abstention or false-premise correction ranges from 6 to 46 responses across the benchmark, and does not track parameter count.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.40996 — 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 2026-11-03T07:29:47.957Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).