Context-Adaptive Synthesis and Compression for Enhanced Retrieval-Augmented Generation in Complex Domains

TL;DR

CASC is a novel framework that enhances retrieval-augmented generation in complex domains by intelligently synthesizing and compressing retrieved information, leading to more accurate and trustworthy answers.

Contribution

We introduce CASC, a context-adaptive synthesis and compression framework with a specialized module for better information processing in complex, multi-document retrieval scenarios.

Findings

CASC outperforms strong baselines on SciDocs-QA dataset.

Significantly reduces token count and cognitive load for LLMs.

Improves accuracy and trustworthiness in complex scientific QA tasks.

Abstract

Large Language Models (LLMs) excel in language tasks but are prone to hallucinations and outdated knowledge. Retrieval-Augmented Generation (RAG) mitigates these by grounding LLMs in external knowledge. However, in complex domains involving multiple, lengthy, or conflicting documents, traditional RAG suffers from information overload and inefficient synthesis, leading to inaccurate and untrustworthy answers. To address this, we propose CASC (Context-Adaptive Synthesis and Compression), a novel framework that intelligently processes retrieved contexts. CASC introduces a Context Analyzer & Synthesizer (CAS) module, powered by a fine-tuned smaller LLM, which performs key information extraction, cross-document consistency checking and conflict resolution, and question-oriented structured synthesis. This process transforms raw, scattered information into a highly condensed, structured, and semantically rich context, significantly reducing the token count and cognitive load for the final Reader LLM. We evaluate CASC on SciDocs-QA, a new challenging multi-document question answering dataset designed for complex scientific domains with inherent redundancies and conflicts. Our extensive experiments demonstrate that CASC consistently outperforms strong baselines.