NeuSym-RAG: Hybrid Neural Symbolic Retrieval with Multiview Structuring for PDF Question Answering

TL;DR

NeuSym-RAG introduces a hybrid neural-symbolic retrieval framework for PDF question answering, leveraging multi-view chunking and schema parsing to improve information retrieval from complex, structured academic papers.

Contribution

It presents a novel interactive retrieval method combining neural and symbolic approaches with multi-view PDF structuring, enhancing QA performance on academic documents.

Findings

Outperforms vector-based RAG and structured baselines on PDF QA datasets.

Effectively utilizes multiple views and structured data for improved retrieval.

Demonstrates stable and superior performance across diverse datasets.

Abstract

The increasing number of academic papers poses significant challenges for researchers to efficiently acquire key details. While retrieval augmented generation (RAG) shows great promise in large language model (LLM) based automated question answering, previous works often isolate neural and symbolic retrieval despite their complementary strengths. Moreover, conventional single-view chunking neglects the rich structure and layout of PDFs, e.g., sections and tables. In this work, we propose NeuSym-RAG, a hybrid neural symbolic retrieval framework which combines both paradigms in an interactive process. By leveraging multi-view chunking and schema-based parsing, NeuSym-RAG organizes semi-structured PDF content into both the relational database and vectorstore, enabling LLM agents to iteratively gather context until sufficient to generate answers. Experiments on three full PDF-based QA datasets, including a self-annotated one AIRQA-REAL, show that NeuSym-RAG stably defeats both the vector-based RAG and various structured baselines, highlighting its capacity to unify both retrieval schemes and utilize multiple views. Code and data are publicly available at https://github.com/X-LANCE/NeuSym-RAG.