Skip to content

LLMs1 min read

Solving versus Verifying: Catching Contradictions in Tax Reasoning Systems

arXiv:2609.05928v1 Announce Type: new Abstract: Large language models now compute correct tax liabilities on over 90% of well-formed cases in statutory benchmarks, which makes them candidates for the tax-advisory and compliance systems t...

By OpenSmartRoute editorial · attributed excerpt

From arXiv cs.CL

arXiv:2609.05928v1 Announce Type: new Abstract: Large language models now compute correct tax liabilities on over 90% of well-formed cases in statutory benchmarks, which makes them candidates for the tax-advisory and compliance systems that consume such an answer directly. Real legal inputs, however, are frequently defective: required facts are missing, or stated facts contradict one another. Accuracy on clean benchmarks says nothing about how a model behaves then, and a system that computes straight through a defective input returns a confident number with no sign that anything is wrong. This raises two questions: does a model asked to solve a case abstain when the input is defective, and when it does not, can the same model catch the defect when asked instead to verify the input? We study six recent models on SARA-derived tax cases under missing-fact and contradictory-fact perturbations. The strongest models abstain when a fact is missing but compute through injected contradictions, returning the clean-input answer 63-76% of the time with no signal of the conflict; asked instead to verify the same input, they flag most of those contradictions. We wire that verification call into a simple contradiction gate: one extra call that abstains when the model reports a conflict. Across all six models it recovers most of the missed contradiction abstention at a clean-accuracy cost of at most about 5 percentage points, with no training and no external tooling. High accuracy on well-formed inputs is therefore an incomplete measure of reliability, and the detection the solver misses is cheaply recoverable with a single self-check.

Read the original at arXiv cs.CL: Solving versus Verifying: Catching Contradictions in Tax Reasoning Systems

Source: https://arxiv.org/abs/2609.05928

Topics

Published Sep 9, 2026 · updated Sep 9, 2026 · 273 words

This post is an attributed excerpt of the source above (unusable model output); no model rewrote it. Refer to the source for the authoritative text.

Keep reading

Related posts

More in LLMs

LLMs1 min read

When Agent Governance Helps

arXiv:2609.05531v1 Announce Type: new Abstract: No specification says how a governed autotelic AI agent organization, where agents pursue self-generated goals inside guardrails, should be designed and evaluated. We answer in two parts. F...

LLMs1 min read

CONDUIT: KV Cache Reuse Policy for VLMs

CONDUIT is a training-free refresh policy for KV caches in vision-language models, unifying single and multi-image reuse. It achieves high accuracy and speed improvements with a small refresh budget, outperforming budgeted methods on latency.