Newest papers

A Neuro-Symbolic Reasoning Engine for Compilation-Free Patch Presence Testing

Siyuan Li, Jiayi Zhou, Zhiyuan Xia, Weijie Wang, Yaowen Zheng, Weiran Tang, Hong Li, Hongsong Zhu, Shanqing Guo · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.27871 · cited by 0 works (OpenAlex) · open-access PDF

Korean

이 코드는 현대 소프트웨어 생태계에서 공식적으로 재사용되므로 1 일 취약점의 확산을 촉진하여 상당한 보안 위험을 초래합니다. 패치가 존재하는지 결정하는 것은 중요하지만 전통적인 방법은 약간의 유용성과 정확성에 의해 고통 받고 종종 취약한 컴파일 프로세스에 의존합니다. 우리의 이전 작업에서 우리는 Lares를 도입했으며, 첫 번째 패치 존재 테스트를위한 컴파일 무료 프레임 워크는 SMT 솔버와 큰 언어 모델 (LLMs)을 결합합니다. 그러나 Lares는 LLM과 SMT 솔버를 두 개의 분리 된 병렬 결정 채널로 취급합니다. 상징적 채널 (Z3)은 코드 사례의 20 %에 성공했으며 나머지 사례는 신경 채

Machine translation (Meta M2M100); the English original follows.

Abstract

Code reuse in modern software ecosystems facilitates the spread of 1-day vulnerabilities, posing significant security risks. Determining if patches are present is crucial, but traditional methods suffer from limited usability and accuracy, often depending on a fragile compilation process. In our prior work, we introduced Lares, the first compilation-free framework for patch presence testing that combines large language models (LLMs) with SMT solvers. However, Lares treats the LLM and the SMT solver as two separate, parallel decision channels: the symbolic channel (Z3) succeeded in only 20% of code cases, and the remaining cases fell back to the neural channel (the LLM) for the final decision. Letting the LLM act as the final arbiter turns verification into an informal, non-deterministic process that is prone to hallucination errors, while the two channels never reinforce each other. To overcome these limitations, we propose SyLares, which reformulates compilation-free patch presence testing as neuro-symbolic reasoning. Instead of letting the LLM emit a final verdict, SyLares uses the LLM as a neural feature extractor that perceives heterogeneous source code and decompiled pseudocode and produces specification-compliant SMT-LIB inputs, and delegates the final, deterministic judgment to the symbolic Z3 solver. This neural-perceives, symbolic-decides coupling is realized by a Code2SMT reasoning engine that combines deterministic syntactic parsing with LLM-driven code normalization and symbol alignment, so that every case is reduced to a formal equivalence query rather than an opaque LLM guess.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Programming language, Symbolic execution, Reuse, Nondeterministic algorithm, Concolic testing

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange