# Reasoners or Translators? Contamination-aware Evaluation and Neuro-Symbolic Robustness in Tax Law

arXiv: https://arxiv.org/abs/2605.16052 · HTML: https://mlsanity.com/paper/2605.16052/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 15 May 2026.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 20 of 20)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning](https://mlsanity.com/paper/2305.12295/) (2023)
- [LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models](https://mlsanity.com/paper/2308.11462/) (2023)
- [LexGLUE: A Benchmark Dataset for Legal Language Understanding in English](https://mlsanity.com/paper/2110.00976/) (2021)
- [Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools](https://mlsanity.com/paper/2405.20362/) (2024)
- [Large Legal Fictions: Profiling Legal Hallucinations in Large Language Models](https://mlsanity.com/paper/2401.01301/) (2024)
- [A Dataset for Statutory Reasoning in Tax Law Entailment and Question Answering](https://mlsanity.com/paper/2005.05257/) (2020)
- [Harnessing the Power of Large Language Models for Natural Language to First-Order Logic Translation](https://mlsanity.com/paper/2305.15541/) (2023)
- [Data Contamination Quiz: A Tool to Detect and Estimate Contamination in Large Language Models](https://mlsanity.com/paper/2311.06233/) (2023)
- [Large Language Models as Tax Attorneys: A Case Study in Legal Capabilities Emergence](https://mlsanity.com/paper/2306.07075/) (2023)
- [Unlocking Practical Applications in Legal Domain: Evaluation of GPT for Zero-Shot Semantic Annotation of Legal Texts](https://mlsanity.com/paper/2305.04417/) (2023)
- [LeanReasoner: Boosting Complex Logical Reasoning with Lean](https://mlsanity.com/paper/2403.13312/) (2024)
- [On Verifiable Legal Reasoning: A Multi-Agent Framework with Formalized Knowledge Representations](https://mlsanity.com/paper/2509.00710/) (2025)
- [PLawBench: A Rubric-Based Benchmark for Evaluating LLMs in Real-World Legal Practice](https://mlsanity.com/paper/2601.16669/) (2026)
- [VeriCoT: Neuro-symbolic Chain-of-Thought Validation via Logical Consistency Checks](https://mlsanity.com/paper/2511.04662/) (2025)
- [Evaluating Test-Time Scaling LLMs for Legal Reasoning: OpenAI o1, DeepSeek-R1, and Beyond](https://mlsanity.com/paper/2503.16040/) (2025)
- [NL2LOGIC: AST-Guided Translation of Natural Language into First-Order Logic with Large Language Models](https://mlsanity.com/paper/2602.13237/) (2026)
- [Language Models and Logic Programs for Trustworthy Tax Reasoning](https://mlsanity.com/paper/2508.21051/) (2025)
- [Reframing Tax Law Entailment as Analogical Reasoning](https://mlsanity.com/paper/2401.06715/) (2024)

## Abstract (arXiv)

Recent advances in large language models (LLMs) have significantly enhanced automated legal reasoning. Yet, it remains unclear whether their performance reflects genuine legal reasoning ability or artifacts of data contamination. We present a comprehensive empirical study of tax law reasoning approaches and implement a contamination detection protocol to rigorously assess LLM reliability. We show that performance can be inflated by contamination. Building on this analysis, we conduct a systematic evaluation, comparing monolithic LLMs with hybrid systems that translate statutory text into formal representations and delegate inference to symbolic solvers. We build a novel test suite designed to probe generalization to unseen documents via case and rule variations. Our findings indicate that legal reasoning is inherently compositional and that neuro-symbolic frameworks offer a more reliable and robust foundation for legal AI, as well as improved generalization to unobserved situations.