Nlp

LLM-based applications face critical security challenges in form of prompt injections and jailbreaks. This article dives into the key architectural improvements underpinning ModernBERT, and demonstrates how to fine-tune it for discriminating malicious prompts. Our model closely approximates the performance of Claude 3.7 and Gemini Flash 2.0 on a mixed benchmark (NotInject, BIPIA, Wildguard-Benign, and PINT), while maintaining low latency (<40ms).

The ever-growing volume of research publications necessitates efficient methods for structuring academic knowledge. We present an end-to-end automated solution using embedding quantization and an LLM pipeline. Our case study starts with a dataset of 25,000 arXiv publications from Computational Linguistics, which we organize under a novel taxonomy of classes.