New Benchmark Tests Bengali-Language Models for Fabricated Islamic Scripture
Researchers introduced BILAL-BENCH, a benchmark measuring how often open language models invent Qur'anic verses or hadith when responding in Bengali, a form of hallucination that lends false authority to sacred texts.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-02 05:00 · arXiv cs.CL
Do Open Language Models Invent Scripture? BILAL-BENCH: Benchmarking Islamic-Source Fabrication in Bengali