A developer found that the bge-reranker-base cross-encoder silently truncates passages at 512 tokens, cutting the tail of 31% of their RAG chunks and causing the reranker to demote correct answers tha…
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…
A new independent evaluation of OpenAI's Privacy Filter (OPF), a 1.5B-parameter PII detector, across 42 benchmarks in 22 languages and 5 domains finds that OPF outperforms Presidio and XLM-RoBERTa on …
Researchers developed a multilingual, multimodal NLP framework for early detection of fake news and violence-driven mob activity, achieving 98% accuracy on a fused dataset of 138,256 Bangla and Englis…
A developer improved a multilingual fact-checker's F1 score from 0.655 to 0.813 by fixing a fundamental input error: the model was trained on claims alone instead of claim-evidence pairs. The XLM-RoBE…
Perplexity AI open-sourced a Unigram tokenizer reimplemented in Rust that achieves 5x lower p50 latency than the Hugging Face tokenizers crate at production input lengths. The new tokenizer, available…