Search Results for author: Tamas Joo

Found 2 papers, 0 papers with code

Investigating the performance of Retrieval-Augmented Generation and fine-tuning for the development of AI-driven knowledge-based systems

no code implementations12 Mar 2024 Robert Lakatos, Peter Pollner, Andras Hajdu, Tamas Joo

Furthermore, we outline a simple RAG-based architecture which, on average, outperforms the FN models by 16% in terms of the ROGUE score, 15% in the case of the BLEU score, and 53% based on the cosine similarity.

Domain Adaptation Hallucination

Cannot find the paper you are looking for? You can Submit a new open access paper.