TIDE, an expert-verified benchmark of 3,050 QA pairs over 644 official customs instruments issued between 1969 and 2025 by the Government of Bangladesh, is presented, covering eight task types over deeply code-mixed documents that are heterogeneous in layout and dated in two calendars.
M. Sobhani, Md. Faiyaz Abdullah Sayeedi, F. Chowdhury et al.· 1 citation· ⚡1
Using a three-phase protocol that separates diagram-only, diagram-plus-description, and description-only inputs, it is shown across five open-weight and closed-source VLMs that description-only performance is statistically indistinguishable from diagram-plus-description performance, establishing structured descriptions...
Khan Raiyan Ibne Reza, Sanjana Aktar Maria, Sumaiya Tabassum Nimi et al.· 0 citations
Fine-tuning can improve legal question-answering accuracy without improving how models use law supplied in context. We study this distinction in bilingual Bangladeshi legal QA, where observed errors can arise from answer scoring, retrieval, or failure to use relevant law. We construct a hierarchy-preserving statutory c...
