Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DunbaaBERT: From Sacrifice to Semantics

About

Large language models have achieved strong performance across many NLP tasks, yet Urdu remains comparatively underexplored due to limited resources and fragmented evaluation settings. To address this gap, we introduce DunbaaBERT, a family of Urdu RoBERTa-base models trained from scratch with Byte-BPE vocabularies of 32k, 52k, and 96k tokens on a deduplicated 17GB Urdu corpus. We evaluate DunbaaBERT across intrinsic and downstream Urdu NLP benchmarks covering linguistic acceptability, news classification, offensive language detection, and sentiment analysis while analyzing vocabulary-size effects on performance and efficiency trade-offs. Across benchmarks, the DunbaaBERT variants achieve competitive performance against strong multilingual baselines while consistently maintaining favorable efficiency trade-offs. Interestingly, larger vocabularies do not consistently improve downstream effectiveness, with DunbaaBERT$_{\text{32k}}$ repeatedly providing the strongest overall efficiency profile. Overall, our results demonstrate that carefully curated Urdu-specific encoder models can remain highly competitive despite comparatively compact model and training scales. All models are released under the MIT license.

Iffat Maab, Waleed Jamil, Raphael Schmitt• 2026

Related benchmarks

TaskDatasetResultRank
Linguistic AcceptabilityURBLIMP
Aspect Accuracy100
10
Offensive Language DetectionUSADC
Macro F10.9408
10
News Domain ClassificationCOUNT 19
Macro-F195.22
10
Sentiment ClassificationPSL–Kabaddi
Macro-F170.53
10
Sentiment ClassificationIMDB Urdu
Macro F190.65
10
Showing 5 of 5 rows

Other info

Follow for update