Imandra CodeLogician: Neuro-Symbolic Reasoning for Precise Analysis of Software Logic
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911428544299008 |
|---|---|
| author | Lin, Hongyu Abdallah, Samer Valentinov, Makar Brennan, Paul Kagan, Elijah Wintersteiger, Christoph M. Ignatovich, Denis Passmore, Grant |
| author_facet | Lin, Hongyu Abdallah, Samer Valentinov, Makar Brennan, Paul Kagan, Elijah Wintersteiger, Christoph M. Ignatovich, Denis Passmore, Grant |
| contents | Large Language Models (LLMs) have shown strong performance on code understanding tasks, yet they fundamentally lack the ability to perform precise, exhaustive mathematical reasoning about program behavior. Existing benchmarks either focus on mathematical proof automation, largely disconnected from real-world software, or on engineering tasks that do not require semantic rigor.
We present CodeLogician, a neurosymbolic agent for precise analysis of software logic, integrated with ImandraX, an industrial automated reasoning engine deployed in financial markets and safety-critical systems. Unlike prior approaches that use formal methods primarily to validate LLM outputs, CodeLogician uses LLMs to construct explicit formal models of software systems, enabling automated reasoning to answer rich semantic questions beyond binary verification outcomes.
To rigorously evaluate mathematical reasoning about software logic, we introduce code-logic-bench, a benchmark targeting the middle ground between theorem proving and software engineering benchmarks. It measures reasoning correctness about program state spaces, control flow, coverage constraints, and edge cases, with ground truth defined via formal modeling and region decomposition.
Comparing LLM-only reasoning against LLMs augmented with CodeLogician, formal augmentation yields substantial improvements, closing a 41-47 percentage point gap in reasoning accuracy. These results demonstrate that neurosymbolic integration is essential for scaling program analysis toward rigorous, autonomous software understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_11840 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Imandra CodeLogician: Neuro-Symbolic Reasoning for Precise Analysis of Software Logic Lin, Hongyu Abdallah, Samer Valentinov, Makar Brennan, Paul Kagan, Elijah Wintersteiger, Christoph M. Ignatovich, Denis Passmore, Grant Artificial Intelligence Logic in Computer Science Software Engineering 68N30 F.3.1; D.2.4; I.2.3; I.2.4 Large Language Models (LLMs) have shown strong performance on code understanding tasks, yet they fundamentally lack the ability to perform precise, exhaustive mathematical reasoning about program behavior. Existing benchmarks either focus on mathematical proof automation, largely disconnected from real-world software, or on engineering tasks that do not require semantic rigor. We present CodeLogician, a neurosymbolic agent for precise analysis of software logic, integrated with ImandraX, an industrial automated reasoning engine deployed in financial markets and safety-critical systems. Unlike prior approaches that use formal methods primarily to validate LLM outputs, CodeLogician uses LLMs to construct explicit formal models of software systems, enabling automated reasoning to answer rich semantic questions beyond binary verification outcomes. To rigorously evaluate mathematical reasoning about software logic, we introduce code-logic-bench, a benchmark targeting the middle ground between theorem proving and software engineering benchmarks. It measures reasoning correctness about program state spaces, control flow, coverage constraints, and edge cases, with ground truth defined via formal modeling and region decomposition. Comparing LLM-only reasoning against LLMs augmented with CodeLogician, formal augmentation yields substantial improvements, closing a 41-47 percentage point gap in reasoning accuracy. These results demonstrate that neurosymbolic integration is essential for scaling program analysis toward rigorous, autonomous software understanding. |
| title | Imandra CodeLogician: Neuro-Symbolic Reasoning for Precise Analysis of Software Logic |
| topic | Artificial Intelligence Logic in Computer Science Software Engineering 68N30 F.3.1; D.2.4; I.2.3; I.2.4 |
| url | https://arxiv.org/abs/2601.11840 |