The BasePeople

Provat Saha

Titles as the sources gave them. In 1 entry, with 0 checked quotations.

In the Base

Every entry they appear in.

  1. ResearcharXivWrote it

    BDBBQ: A Human-AI Collaborative Benchmark for Bengali Social Bias in LLMs

    According to the abstract, Provat Saha and colleagues build BDBBQ, a bilingual benchmark of 61,312 question-answering instances across 12 social categories to measure bias in language models in Bengali and English. Eight models were evaluated; the abstract reports lower accuracy in Bengali for all of them.