Enabling Equitable Access to Trustworthy Financial Reasoning

2025-08-28

AI Summary

This work addresses the challenges of complex tax regulations, high error penalties, and insufficient trustworthiness and accessibility of automated tax-filing systems. We propose a neuro-symbolic architecture integrating large language models (LLMs) with symbolic solvers. Our core innovation lies in pre-translating natural-language tax rules into executable logic programs and augmenting reasoning with case-based retrieval to enhance both accuracy and auditability. Experiments on the SARA dataset demonstrate substantial improvements in reasoning accuracy while reducing deployment costs to a fraction of typical industry benchmarks. Notably, we introduce the first economic feasibility analysis grounded in real-world tax penalty costs. The framework achieves high precision and strong interpretability without compromising fairness, accessibility, or practical deployability—marking a significant step toward real-world adoption of neuro-symbolic systems in mission-critical public-service domains.

Technology Category

• Machine Learning: Neuro-Symbolic Learning
• Cognitive Modeling & Cognitive Systems: Symbolic Representations
• Natural Language Processing: Safety and Robustness

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

According to the United States Internal Revenue Service, ''the average American spends $$270$ and 13 hours filing their taxes''. Even beyond the U.S., tax filing requires complex reasoning, combining application of overlapping rules with numerical calculations. Because errors can incur costly penalties, any automated system must deliver high accuracy and auditability, making modern large language models (LLMs) poorly suited for this task. We propose an approach that integrates LLMs with a symbolic solver to calculate tax obligations. We evaluate variants of this system on the challenging StAtutory Reasoning Assessment (SARA) dataset, and include a novel method for estimating the cost of deploying such a system based on real-world penalties for tax errors. We further show how combining up-front translation of plain-text rules into formal logic programs, combined with intelligently retrieved exemplars for formal case representations, can dramatically improve performance on this task and reduce costs to well below real-world averages. Our results demonstrate the promise and economic feasibility of neuro-symbolic architectures for increasing equitable access to reliable tax assistance.

Problem

• Automating complex tax filing with high accuracy
• Integrating LLMs with symbolic solvers for tax calculations
• Reducing costs and errors in tax assistance access

Innovation

• Integrates LLMs with symbolic solver
• Translates rules into formal logic programs
• Uses intelligently retrieved exemplars