2026 · Preprint
A Neuro-Symbolic Reasoning Engine for Compilation-Free Patch Presence Testing
SyLares converts source code and decompiled pseudocode into SMT-LIB equivalence queries, using an LLM to align code and Z3 to decide whether a patch is present.
Scope and limitations
The scope is patch-presence testing across source code and decompiled pseudocode. The reported 20% symbolic-channel success rate concerns the earlier Lares framework.
Why it fits this collection
Its method reasons about behavior represented in decompiled pseudocode, a form of recovered binary code.
Cite
@misc{paper-sylares,
title = {{A Neuro-Symbolic Reasoning Engine for Compilation-Free Patch Presence Testing}},
author = {{Siyuan Li} and {Jiayi Zhou} and {Zhiyuan Xia} and {Weijie Wang} and {Yaowen Zheng} and {Weiran Tang} and {Hong Li} and {Hongsong Zhu} and {Shanqing Guo}},
year = {2026},
url = {https://arxiv.org/abs/2610.27871},
howpublished = {arXiv:2610.27871},
note = {Preprint},
eprint = {2610.27871},
archivePrefix = {arXiv},
month = oct
}
Read in a terminal: hf papers read 2610.27871
Publication and review details
- Venue
- arXiv
- Status
- Preprint
- Publication / posting date
- October 20, 2026
- Date source
- First arXiv submission date.
- First arXiv submission
- October 20, 2026
- arXiv
- 2610.27871
- Reviewed arXiv version
- v1
The date and venue come from the arXiv record.
Sources reviewed
Reviewed against the paper and its primary sources. Cite the original authors using the linked paper or BibTeX.