Fin-STAR: Structure-as-Semantics to Resolve Implicitness in Financial Retrieval

ACL ARR 2026 January Submission5855 Authors

Keywords: Retrieval-Augmented Generation, Financial Document Understanding, Document Structure Understanding, Multi-hop Reasoning, Semantic Implicitness

Abstract: Understanding financial documents is critical for high-stakes decision-making yet hindered by systemic semantic implicitness: key facts are rarely explicit in surface text and often determined by global structural cues. Missing these cues invites semantic misinterpretations, such as misreading what a number refers to, an outcome unacceptable in high-stakes environments. However, existing Retrieval-Augmented Generation (RAG) systems typically treat structure as a physical navigational skeleton rather than intrinsic semantic knowledge. To address this, we introduce Fin-STAR (Financial STructure-As-Semantics Retrieval), a framework redefining hierarchy as intrinsic semantics. Fin-STAR incorporates a novel Structure-Enriched Semantic Indexing mechanism that augments the hierarchical lineage with snippet-derived virtual nodes, and injects this enriched context via a semantic cross-attention paradigm, rendering implicit cues explicit. By grounding evidence within its structural scope, we preserve factual invariance and ensure contextual integrity. Addressing the lack of granular public datasets, we conduct experiments on FinTierQA Gold, a curated expert benchmark. Results show that Fin-STAR outperforms state-of-the-art hierarchical and graph-based baselines across diverse query complexities, document types, and markets. Notably, ablations confirm that our semantic injection consistently outperforms alternative strategies. Finally, we release FinTierQA, comprising 3.9M pairs automatically constructed from 78k documents via our framework .

Paper Type: Long

Research Area: Retrieval-Augmented Language Models

Research Area Keywords: retrieval-augmented generation, financial/business NLP, logical reasoning, multihop QA

Contribution Types: NLP engineering experiment, Data resources

Languages Studied: English, Simplified Chinese, Traditional Chinese

Submission Number: 5855