# Humans vs. Machines: A Comparison With AI Agents in Software Reverse Engineering

> Compares human and LLM-agent understanding and navigation on the same static reverse-engineering challenges, then studies how binary representations consume context windows.

- Authors: Samuele Doria; Stefano De Rosa; Ruoyu Wang; Zion Leonahenahe Basque; Eleonora Losiouk; Simone Aonzo
- Venue: arXiv
- Status: Preprint
- Publication / posting date: October 10, 2026
- Date source: arXiv v1 submission
- First arXiv submission: October 10, 2026
- arXiv: 2610.14960
- Reviewed arXiv version: v1
- DOI: 10.48550/arxiv.2610.14960
- Sources reviewed: 2026-10-16
- Added to the catalog: 2026-10-16
- Original paper: <https://arxiv.org/abs/2610.14960>
- PDF: <https://arxiv.org/pdf/2610.14960>
- BibTeX: <https://awesomerea.org/papers/citations/paper-humans-machines-sre.bib>
- Catalog page: <https://awesomerea.org/papers/paper-humans-machines-sre/>
- Record JSON: <https://awesomerea.org/papers/paper-humans-machines-sre/record.json>
- Chinese version: [简体中文](https://awesomerea.org/zh/papers/paper-humans-machines-sre/index.md)

## Scope and limitations

Understanding is graded on two compact challenges; the separate 4,000-binary study measures context capacity, not task-solving accuracy. The interface is static IDA/MCP without dynamic execution, and prior exposure to challenge source remains an acknowledged validity threat.

## Why it fits this collection

Informs tool-interface design, selective retrieval, and persistent analysis state for reverse-engineering agents.

## Publication and review notes

The original source review did not verify a conference venue or paper-specific public code link. On 2026-10-20, the official arXiv history still listed only v1. Although its abstract page labels the DOI as pending, DataCite reports the matching title and six authors as findable, registered on 2026-10-14. This is an identifier correction, not a new paper version or expanded experimental verification.

## Primary sources

1. <https://arxiv.org/abs/2610.14960>
2. <https://arxiv.org/html/2610.14960v1>
3. <https://api.datacite.org/dois/10.48550/arxiv.2610.14960>

## About this entry

Reviewed against the paper and its primary sources. Cite the original authors using the linked paper or BibTeX.
Original editorial content: MIT. Linked projects, papers and assets retain their own rights and licenses.