{"ID":2857147,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.10252","arxiv_id":"2510.10252","title":"Audit-of-Understanding: Posterior-Constrained Inference for Mathematical Reasoning in Language Models","abstract":"Large language models (LLMs) often generate reasoning traces that appear coherent but rest on unsupported assumptions, leading to hallucinated conclusions. Prior work mainly addresses factual hallucinations or relies on post-hoc verification, leaving reasoning-induced hallucinations largely unaddressed. We propose Audit-of-Understanding (AoU), a framework that constrains inference to validated premises through three phases: (1) decomposing a query into candidate assumptions, (2) auditing their support, and (3) conditioning inference only on the validated subset. Formally, AoU is \\emph{posterior-constrained inference}, connecting to selective prediction and rejection learning. Our contributions are threefold: (i) theoretical guarantees under perfect validation, (ii) excess-risk bounds under imperfect audits, and (iii) tractability analysis. Empirically, AoU improves both accuracy and faithfulness on GSM8K, MultiArith, and SVAMP, achieving up to +30% gains on GSM8K, +45% on MultiArith, and consistent +20--28% improvements on SVAMP over Chain-of-Thought, Self-Consistency, and CoT-Decoding. Code is available at https://anonymous.4open.science/r/audit-of-understanding-E28B.","short_abstract":"Large language models (LLMs) often generate reasoning traces that appear coherent but rest on unsupported assumptions, leading to hallucinated conclusions. Prior work mainly addresses factual hallucinations or relies on post-hoc verification, leaving reasoning-induced hallucinations largely unaddressed. We propose Audi...","url_abs":"https://arxiv.org/abs/2510.10252","url_pdf":"https://arxiv.org/pdf/2510.10252v2","authors":"[\"Samir Abdaljalil\",\"Erchin Serpedin\",\"Khalid Qaraqe\",\"Hasan Kurban\"]","published":"2025-10-11T15:10:28Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","project_urls":"[\"https://anonymous.4open.science/r/audit-of-understanding-E28B\"]","has_code":false}