Ingest-And-Ground: Dispelling Hallucinations from Continually-Pretrained LLMs with RAG

Ingest-And-Ground: Dispelling Hallucinations from Continually-Pretrained LLMs with RAG

This paper presents new methods that have the potential to improve privacy process efficiency with LLM and RAG. To reduce hallucination, we continually pre-train the base LLM model with a privacy-specific knowledge base and then augment it with a semantic RAG layer. Our evaluations demonstrate that...

Full description

Saved in:
Bibliographic Details
Main Authors: Fang, Chenhao, Larson, Derek, Zhu, Shitong, Zeng, Sophie, Summer, Wendy, Peng, Yanqing, Hulovatyy, Yuriy, Rao, Rajeev, Forgues, Gabriel, Pudota, Arya, Goncalves, Alex, Robert, Hervé
Format: Journal Article
Language:English
Published: 30-09-2024
Subjects:
Computer Science - Computation and Language
Computer Science - Cryptography and Security
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents new methods that have the potential to improve privacy process efficiency with LLM and RAG. To reduce hallucination, we continually pre-train the base LLM model with a privacy-specific knowledge base and then augment it with a semantic RAG layer. Our evaluations demonstrate that this approach enhances the model performance (as much as doubled metrics compared to out-of-box LLM) in handling privacy-related queries, by grounding responses with factual information which reduces inaccuracies.
DOI:10.48550/arxiv.2410.02825