Fine-tuned XLM-RoBERTa achieved an F1-score of 0.5959 on Bangla medical named entity recognition, setting a new state of the art above the previously reported best of 0.5848, according to an arXiv ben…
Researchers proposed BanglaMamba, a Mamba-based State Space Model for Bangla fake news detection, and found it achieves a Macro-F1 score of 0.9029, comparable to a from-scratch CustomBERT (0.9057) but…
Researchers introduced destroR, a pipeline with three attack strategies—paraphrase, back-translation, and one-hot word-swap—that revealed Bangla sentiment classifiers are vulnerable to adversarial att…
Researchers introduced a Bangla event detection benchmark with 9,979 annotated sentences across clean text, ASR transcripts, and corrupted text. They found encoder models perform better on clean text …