LettuceDetect: A Hallucination Detection Framework for RAG Applications

By Adam Kovacs, Gábor Recski. 2025

Subfields: Retrieval-Augmented Generation, Evaluation & Benchmarks

Citations in the corpus

• Turk-LettuceDetect: A Hallucination Detection Model for Turkish RAG Applications (2025, unknown): “Notable examples include Luna [10], which uses a classification head over contextual embeddings to identify unsupported content, and LettuceDetect [1], which extends this idea with token-level classification and achieves significant improvements in precision and recall over prior encoder-based…”