Reza Kazemi

FaMTEB: Massive Text Embedding Benchmark in Persian Language

2025 · arXiv (Cornell University)

In this paper, we introduce a comprehensive benchmark for Persian (Farsi) text embeddings, built upon the Massive Text Embedding Benchmark (MTEB). Our benchmark includes 63 datasets spanning seven different tasks: classification, clustering, pair classification, reranking, …

Enhancing Adversarial Robustness With Energy-Based Models: a Novel Training Approach for Input Purification

2025

Deep learning has advanced in fields like computer vision and natural language processing but still faces reliability challenges, including adversarial attacks that mislead models and threaten security-sensitive applications. Generative models help counter such threats by …