# Decomposing Retrieval Failures in RAG for Long-Document Financial Question Answering

arXiv: https://arxiv.org/abs/2602.17981 · HTML: https://mlsanity.com/paper/2602.17981/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Legal and financial RAG, one of 15 directions in Retrieval-augmented generation.
- Builds on 18 of its 23 references that are in the map.
- First posted on arXiv 20 Feb 2026.

## Placed in

- Retrieval-augmented generation / [Legal and financial RAG](https://mlsanity.com/rag/legal-and-financial-rag/)

## Builds on (most cited first, 18 of 18)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering](https://mlsanity.com/paper/2007.01282/) (2020)
- [M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation](https://mlsanity.com/paper/2402.03216/) (2024)
- [FinQA: A Dataset of Numerical Reasoning over Financial Data](https://mlsanity.com/paper/2109.00122/) (2021)
- [Precise Zero-Shot Dense Retrieval without Relevance Labels](https://mlsanity.com/paper/2212.10496/) (2022)
- [TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance](https://mlsanity.com/paper/2105.07624/) (2021)
- [SPLADE: Sparse Lexical and Expansion Model for First Stage Ranking](https://mlsanity.com/paper/2107.05720/) (2021)
- [FinanceBench: A New Benchmark for Financial Question Answering](https://mlsanity.com/paper/2311.11944/) (2023)
- [Efficient Natural Language Response Suggestion for Smart Reply](https://mlsanity.com/paper/1705.00652/) (2017)
- [ConvFinQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answering](https://mlsanity.com/paper/2210.03849/) (2022)
- [Financial Report Chunking for Effective Retrieval Augmented Generation](https://mlsanity.com/paper/2402.05131/) (2024)
- [FinSage: A Multi-aspect RAG System for Financial Filings Question Answering](https://mlsanity.com/paper/2504.14493/) (2025)
- [SEC-QA: A Systematic Evaluation Corpus for Financial QA](https://mlsanity.com/paper/2406.14394/) (2024)
- [Optimizing Retrieval Strategies for Financial Question Answering Documents in Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2503.15191/) (2025)
- [Hierarchical Retrieval with Evidence Curation for Open-Domain Financial Question Answering on Standardized Documents](https://mlsanity.com/paper/2505.20368/) (2025)
- [FinGEAR: Financial Mapping-Guided Enhanced Answer Retrieval](https://mlsanity.com/paper/2509.12042/) (2025)
- [Enhancing Financial RAG with Agentic AI and Multi-HyDE: A Novel Approach to Knowledge Retrieval and Hallucination Reduction](https://mlsanity.com/paper/2509.16369/) (2025)

## Abstract (arXiv)

Retrieval-augmented generation is increasingly used for financial question answering over long regulatory filings, yet reliability depends on retrieving the exact context needed to justify answers in high stakes settings. We study a frequent failure mode in which the correct document is retrieved but the page or chunk that contains the answer is missed, leading the generator to extrapolate from incomplete context. Despite its practical significance, this within-document retrieval failure mode has received limited systematic attention in the Financial Question Answering (QA) literature. We evaluate retrieval at multiple levels of granularity, document, page, and chunk level, and introduce an oracle based analysis to provide empirical upper bounds on retrieval and generative performance. On a 150 question subset of FinanceBench, we reproduce and compare diverse retrieval strategies including dense, sparse, hybrid, and hierarchical methods with reranking and query reformulation. Across methods, gains in document discovery tend to translate into stronger page recall, yet oracle performance still suggests headroom for page and chunk level retrieval. To target this gap, we introduce a domain fine-tuned page scorer that treats pages as an intermediate retrieval unit between documents and chunks. Unlike prior passage-based hierarchical retrieval, we fine-tune a bi-encoder specifically for page-level relevance on financial filings, exploiting the semantic coherence of pages. Overall, our results demonstrate a significant improvement in page recall and chunk retrieval.