Fighting Numerical Hallucinations via Data-centric Compilation for Online Financial QA

• URL: http://arxiv.org/abs/2605.31064v1
• Date: Fri, 29 May 2026 09:35:11 GMT
• Title: Fighting Numerical Hallucinations via Data-centric Compilation for Online Financial QA
• Abstract summary: Large Language Models (LLMs) have significantly advanced online data services, particularly in the domain of financial question answering (FinQA)<n>However, such systems remain susceptible to numerical reasoning hallucinations, which critically undermine reliability in high-stakes financial applications.<n>We propose a data-centric paradigm and propose a novel framework, the Data-centric Reasoning Compiler (DCRC)<n>We conduct extensive experiments on established offline benchmarks and further validate our framework through deployment in a real-world online financial QA system.
• Score: 14.141735278960766
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large Language Models (LLMs) have significantly advanced online data services, particularly in the domain of financial question answering (FinQA). However, such systems remain susceptible to numerical reasoning hallucinations, which critically undermine reliability in high-stakes financial applications. Although retrieval-augmented generation (RAG) has been widely adopted to ground responses in external knowledge, it introduces three persistent challenges: noise sensitivity, calculation fragility, and an auditability crisis. Existing model-centric approaches, which primarily focus on optimizing either the retriever or generator in isolation, still struggle to address these issues in an integrated manner. In this work, we pioneer a data-centric paradigm and propose a novel framework, the Data-centric Reasoning Compiler (DCRC). The framework operates through three cohesive phases: (1) adversarial data construction, which synthesizes training examples with controlled noise to teach robustness; (2) multi-stage training that cultivates a Data-centric Structuring Agent (DSA) capable of explicit evidence auditing and program synthesis; and (3) a compile-and-execute inference process, where the DSA transforms user queries and retrieved documents into verifiable, executable reasoning programs. This data-driven framework ensures faithful numerical reasoning by design. We conduct extensive experiments on established offline benchmarks and further validate our framework through deployment in a real-world online financial QA system.

Related papers

• VeriTrip: A Verifiable Benchmark for Travel Planning Agents over Unstructured Web Corpora [24.484384823423216]
  Existing benchmarks fail to account for information noise, ignore multi-source factual contradictions, and overlook the necessity of grounding visual perception into logical planning.<n>We introduce VeriTrip, a verifiable benchmark designed to meet the increasing demands for agent robustness and reliability.
  arXiv  Detail & Related papers  (2026-05-27T16:14:47Z)
• FERA: Uncertainty-Aware Federated Reasoning for Large Language Models [60.52562148874846]
  We study federated reasoning, where a server improves multi-step reasoning by coordinating with heterogeneous clients holding private demonstrations.<n>The key challenge is that client reliability is query-dependent, while the server cannot inspect client data to determine which contributions are trustworthy.<n>We propose Uncertainty-Aware Federated Reasoning (FERA), a training-free framework based on iterative server-client co-refinement.
  arXiv  Detail & Related papers  (2026-05-11T07:04:51Z)
• Dual-Loop Control in DCVerse: Advancing Reliable Deployment of AI in Data Centers via Digital Twins [9.44096782663678]
  Deep Reinforcement Learning (DRL) shows strong potential for intelligent control, but its deployment in mission-critical systems is limited by data scarcity and the lack of real-time pre-evaluation mechanisms.<n>This paper introduces the Dual-Loop Control Framework (DLCF), a digital twin-based architecture designed to overcome these challenges.<n>The framework comprises three core entities: the physical system, a digital twin, and a policy reservoir of diverse DRL agents.
  arXiv  Detail & Related papers  (2026-04-08T20:01:34Z)
• Leveraging LLM Parametric Knowledge for Fact Checking without Retrieval [60.25608870901428]
  Trustworthiness is a core research challenge for agentic AI systems built on Large Language Models (LLMs)<n>We propose the task of fact-checking without retrieval, focusing on the verification of arbitrary natural language claims, independent of their source robustness.
  arXiv  Detail & Related papers  (2026-03-05T18:42:51Z)
• Bridging the Arithmetic Gap: The Cognitive Complexity Benchmark and Financial-PoT for Robust Financial Reasoning [11.522192050185568]
  Large Language Models suffer from "Arithmetic Hallucinations" and a systemic failure mode we term "Cognitive Collapse"<n>We introduce the Cognitive Complexity Benchmark (CCB), a robust evaluation framework grounded in a dataset constructed from 95 real-world Chinese A-share annual reports.<n>We propose the Iterative Dual-Phase Financial-PoT framework to address these failures.
  arXiv  Detail & Related papers  (2026-01-29T01:33:33Z)
• VeritasFi: An Adaptable, Multi-tiered RAG Framework for Multi-modal Financial Question Answering [31.334698403426245]
  Retrieval-Augmented Generation (RAG) is becoming increasingly essential for Question Answering (QA) in the financial sector.<n>We present VeritasFi, an innovative hybrid RAG framework that incorporates a multi-modal preprocessing pipeline.<n>By integrating our proposed designs, VeritasFi presents a groundbreaking framework that greatly enhances the adaptability and robustness of financial RAG systems.
  arXiv  Detail & Related papers  (2025-10-12T22:45:24Z)
• CoDA: Agentic Systems for Collaborative Data Visualization [57.270599188947294]
  Deep research has revolutionized data analysis, yet data scientists still devote substantial time to manually crafting visualizations.<n>Existing approaches, including simple single- or multi-agent systems, often oversimplify the task.<n>We introduce CoDA, a multi-agent system that employs specialized LLM agents for metadata analysis, task planning, code generation, and self-reflection.
  arXiv  Detail & Related papers  (2025-10-03T17:30:16Z)
• Learning to Route: A Rule-Driven Agent Framework for Hybrid-Source Retrieval-Augmented Generation [55.47971671635531]
  Large Language Models (LLMs) have shown remarkable performance on general Question Answering (QA)<n>Retrieval-Augmented Generation (RAG) addresses this limitation by enriching LLMs with external knowledge.<n>Existing systems primarily rely on unstructured documents, while largely overlooking relational databases.
  arXiv  Detail & Related papers  (2025-09-30T22:19:44Z)
• Divide-Then-Align: Honest Alignment based on the Knowledge Boundary of RAG [51.120170062795566]
  We propose Divide-Then-Align (DTA) to endow RAG systems with the ability to respond with "I don't know" when the query is out of the knowledge boundary.<n>DTA balances accuracy with appropriate abstention, enhancing the reliability and trustworthiness of retrieval-augmented systems.
  arXiv  Detail & Related papers  (2025-05-27T08:21:21Z)
• MIBP-Cert: Certified Training against Data Perturbations with Mixed-Integer Bilinear Programs [50.41998220099097]
  Data errors, corruptions, and poisoning attacks during training pose a major threat to the reliability of modern AI systems.<n>We introduce MIBP-Cert, a novel certification method based on mixed-integer bilinear programming (MIBP)<n>By computing the set of parameters reachable through perturbed or manipulated data, we can predict all possible outcomes and guarantee robustness.
  arXiv  Detail & Related papers  (2024-12-13T14:56:39Z)
• Soundness of Data-Aware Processes with Arithmetic Conditions [8.914271888521652]
  Data Petri nets (DPNs) have gained increasing popularity thanks to their ability to balance simplicity with expressiveness. The interplay of data and control-flow makes checking the correctness of such models, specifically the well-known property of soundness, crucial and challenging. We provide a framework for assessing soundness of DPNs enriched with arithmetic data conditions.
  arXiv  Detail & Related papers  (2022-03-28T14:46:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.