FinGEAR: Financial Mapping-Guided Enhanced Answer Retrieval

2025/09/15 by Ying Li, Mengyu Wang, Li, Ying +7 · 1 citation
Computer Science · #Advanced Text Analysis Techniques #Computation and Language (cs.CL) #Computational Engineering #FOS: Computer and information sciences #Finance #Information Retrieval and Search Behavior #Topic Modeling #and Science (cs.CE)

paper · pdf · doi:10.48550/arxiv.2509.12042

openalex publication_date 2025/09/15 · openalex created_date 2025/10/12 · openalex updated_date 2026/07/28

Abstract

Financial disclosures such as 10-K filings present challenging retrieval problems due to their length, regulatory section hierarchy, and domain-specific language, which standard retrieval-augmented generation (RAG) models underuse. We introduce FinGEAR (Financial Mapping-Guided Enhanced Answer Retrieval), a retrieval framework tailored to financial documents. FinGEAR combines a finance lexicon for Item-level guidance (FLAM), dual hierarchical indices for within-Item search (Summary Tree and Question Tree), and a two-stage cross-encoder reranker. This design aligns retrieval with disclosure structure and terminology, enabling fine-grained, query-aware context selection. Evaluated on full 10-Ks with queries aligned to the FinQA dataset, FinGEAR delivers consistent gains in precision, recall, F1, and relevancy, improving F1 by up to 56.7% over flat RAG, 12.5% over graph-based RAGs, and 217.6% over prior tree-based systems, while also increasing downstream answer accuracy with a fixed reader. By jointly modeling section hierarchy and domain lexicon signals, FinGEAR improves retrieval fidelity and provides a practical foundation for high-stakes financial analysis.

Citations

• FinBERT-QA: Financial Question Answering with pre-trained BERT Language Models
• INVESTORBENCH: A Benchmark for Financial Decision-Making Tasks with LLM-based Agent
• Modeling News Interactions and Influence for Financial Market Prediction
• In Defense of RAG in the Era of Long-Context Language Models
• The Llama 3 Herd of Models
• BM25S: Orders of magnitude faster lexical search via eager sparse scoring
• Mix-of-Granularity: Optimize the Chunking Granularity for Retrieval-Augmented Generation
• From Local to Global: A Graph RAG Approach to Query-Focused Summarization
• Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity
• RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval
• DocFinQA: A Long-Context Financial Reasoning Dataset
• Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
• Lost in the Middle: How Language Models Use Long Contexts
• BloombergGPT: A Large Language Model for Finance
• GPT-4 Technical Report
• In-Context Retrieval-Augmented Language Models
• Matryoshka Representation Learning
• FinQA: A Dataset of Numerical Reasoning over Financial Data
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• REALM: Retrieval-Augmented Language Model Pre-Training
• ELI5: Long Form Question Answering
• Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
• Replace or Retrieve Keywords In Documents at Scale
• HiQA: A Hierarchical Contextual Augmentation RAG for Multi-Documents QA

Cited by

• One More Question is Enough, Expert Question Decomposition (EQD) Model for Domain Quantitative Reasoning

Related

• Optimizing Retrieval Strategies for Financial Question Answering Documents in Retrieval-Augmented Generation Systems
• Metadata-Driven Retrieval-Augmented Generation for Financial Question Answering
• FinSage: A Multi-aspect RAG System for Financial Filings Question Answering
• Multi-Reranker: Maximizing performance of retrieval-augmented generation in the FinanceRAG challenge
• FinAgentBench: A Benchmark Dataset for Agentic Retrieval in Financial Question Answering