Resolving the Robustness-Precision Trade-off in Financial RAG through Hybrid Document-Routed Retrieval

2026-03-26

AI Summary

Financial Retrieval-Augmented Generation (RAG) systems often struggle to balance robustness and precision when processing structurally homogeneous regulatory documents: passage-level retrieval is susceptible to cross-document confusion, while whole-document routing, though robust, yields coarse-grained localization. To address this, this work proposes a Hybrid Document Routing Retrieval (HDRR) architecture that innovatively integrates Semantic File Routing (SFR) with Chunk-Based Retrieval (CBR) in a two-stage pipeline—first filtering relevant documents and then precisely locating answers—thereby effectively mitigating cross-document interference while preserving fine-grained answer accuracy. Evaluated on the FinDER benchmark, HDRR achieves an average score of 7.54, a failure rate of 6.4%, an accuracy of 67.7%, and a perfect-answer rate of 20.1%, significantly outperforming existing approaches.

Abstract

Retrieval-Augmented Generation (RAG) systems for financial document question answering typically follow a chunk-based paradigm: documents are split into fragments, embedded into vector space, and retrieved via similarity search. While effective in general settings, this approach suffers from cross-document chunk confusion in structurally homogeneous corpora such as regulatory filings. Semantic File Routing (SFR), which uses LLM structured output to route queries to whole documents, reduces catastrophic failures but sacrifices the precision of targeted chunk retrieval. We identify this robustness-precision trade-off through controlled evaluation on the FinDER benchmark (1,500 queries across five groups): SFR achieves higher average scores (6.45 vs. 6.02) and fewer failures (10.3% vs. 22.5%), while chunk-based retrieval (CBR) yields more perfect answers (13.8% vs. 8.5%). To resolve this trade-off, we propose Hybrid Document-Routed Retrieval (HDRR), a two-stage architecture that uses SFR as a document filter followed by chunk-based retrieval scoped to the identified document(s). HDRR eliminates cross-document confusion while preserving targeted chunk precision. Experimental results demonstrate that HDRR achieves the best performance on every metric: an average score of 7.54 (25.2% above CBR, 16.9% above SFR), a failure rate of only 6.4%, a correctness rate of 67.7% (+18.7 pp over CBR), and a perfect-answer rate of 20.1% (+6.3 pp over CBR, +11.6 pp over SFR). HDRR resolves the trade-off by simultaneously achieving the lowest failure rate and the highest precision across all five experimental groups.

Problem

• Robustness-Precision Trade-off
• Financial RAG
• Document-Routed Retrieval
• Chunk-based Retrieval
• Cross-document Confusion

Innovation

• Hybrid Document-Routed Retrieval
• Semantic File Routing
• chunk-based retrieval
• robustness-precision trade-off
• financial RAG