Context-Driven Index Trimming: A Data Quality Perspective to Enhancing Precision of RALMs

TL;DR

This paper introduces Context-Driven Index Trimming (CDIT), a framework that improves RALMs' answer precision by using logical data quality rules to filter inconsistent retrieval results, enhancing overall response accuracy.

Contribution

The paper proposes a novel CDIT framework employing Context Matching Dependencies to regulate data quality in retrieval results for RALMs, a new approach to improve answer precision.

Findings

CDIT effectively filters inconsistent retrieval results.

CDIT improves answer quality across various language models.

The framework is compatible with multiple indexing methods.

Abstract

Retrieval-Augmented Large Language Models (RALMs) have made significant strides in enhancing the accuracy of generated responses.However, existing research often overlooks the data quality issues within retrieval results, often caused by inaccurate existing vector-distance-based retrieval methods.We propose to boost the precision of RALMs' answers from a data quality perspective through the Context-Driven Index Trimming (CDIT) framework, where Context Matching Dependencies (CMDs) are employed as logical data quality rules to capture and regulate the consistency between retrieved contexts.Based on the semantic comprehension capabilities of Large Language Models (LLMs), CDIT can effectively identify and discard retrieval results that are inconsistent with the query context and further modify indexes in the database, thereby improving answer quality.Experiments demonstrate on challenging question-answering tasks.Also, the flexibility of CDIT is verified through its compatibility with various language models and indexing methods, which offers a promising approach to bolster RALMs' data quality and retrieval precision jointly.