{"ID":2870600,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.13471","arxiv_id":"2509.13471","title":"An LLM Agentic Approach for Legal-Critical Software: A Case Study for Tax Prep Software","abstract":"Large language models (LLMs) show promise for translating natural-language statutes into executable logic, but reliability in legally critical settings remains challenging due to ambiguity and hallucinations. We present an agentic approach for developing legal-critical software, using U.S. federal tax preparation as a case study. The key challenge is test-case generation under the oracle problem, where correct outputs require interpreting law. Building on metamorphic testing, we introduce higher-order metamorphic relations that compare system outputs across structured shifts among similar individuals. Because authoring such relations is tedious and error-prone, we use an LLM-driven, role-based framework to automate test generation and code synthesis. We implement a multi-agent system that translates tax code into executable software and incorporates a metamorphic-testing agent that searches for counterexamples. In experiments, our framework using a smaller model (GPT-4o-mini) achieves a worst-case pass rate of 45%, outperforming frontier models (GPT-4o and Claude 3.5, 9-15%) on complex tax-code tasks. These results support agentic LLM methodologies as a path to robust, trustworthy legal-critical software from natural-language specifications.","short_abstract":"Large language models (LLMs) show promise for translating natural-language statutes into executable logic, but reliability in legally critical settings remains challenging due to ambiguity and hallucinations. We present an agentic approach for developing legal-critical software, using U.S. federal tax preparation as a...","url_abs":"https://arxiv.org/abs/2509.13471","url_pdf":"https://arxiv.org/pdf/2509.13471v2","authors":"[\"Sina Gogani-Khiabani\",\"Ashutosh Trivedi\",\"Diptikalyan Saha\",\"Saeid Tizpaz-Niari\"]","published":"2025-09-16T19:13:26Z","proceeding":"cs.SE","tasks":"[\"cs.SE\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}