About পরিচিতি
বাংলাNLP Hub is a community-maintained catalog of research resources for Bengali — the world's sixth most spoken language, and still chronically under-resourced in NLP. We index papers, datasets, pretrained models, tools, and benchmark results in one place so that researchers spend less time hunting and more time building.
The catalog is maintained by ELITE Research Lab with contributions from the wider Bangla NLP community. All data lives in a public GitHub repository; every link is re-verified nightly by CI, and the "✓ verified" chip next to each resource shows when it last resolved.
It currently indexes 813 papers, 63 datasets, 20 models, and 9 tools across 26 tasks. Entries seeded from the original design prototype are being audited against ACL Anthology, arXiv, OpenAlex, and publisher records; several turned out to be inaccurate and were corrected or removed. Dataset size, license, and year have each had an audit pass. Unresolved fields remain listed in the repository.
Corrections, additions, and new benchmark results are welcome — see how to contribute.
Principles
CI re-checks every external link nightly. Failures are collected in a single tracking issue, and the "✓ verified" chip on each resource shows when a human last confirmed it.
Nothing is estimated. Leaderboard rows require a citation to the paper reporting that exact number, so benchmarks with no curated rows show an empty state rather than a plausible guess.
No trackers, no client framework. A handful of small vanilla scripts, built to load fast on mid-range phones and slow connections.
Cite this catalog
If the Hub helped your research, cite it:
@misc{banglanlphub2026,
title = {{B}angla{NLP} Hub: A Community-Maintained Catalog of {B}angla {NLP} Resources},
author = {{ELITE Research Lab} and contributors},
year = {2026},
howpublished = {\url{https://kishormorol.github.io/BanglaNLP-Hub/}},
note = {Accessed 2026-09-16}
}