# HELIOS: Hierarchical Graph Abstraction for Structure-Aware LLM Decompilation

> HELIOS gives an LLM a hierarchical summary of binary control flow and function calls alongside decompiler output, with optional compiler feedback. It evaluates compilability and functional correctness.

- Authors: Yonatan Gizachew Achamyeleh; Harsh Thomare; Mohammad Abdullah Al Faruque
- Venue: arXiv
- Status: Preprint
- Publication / posting date: January 21, 2026
- Date source: First arXiv submission date.
- First arXiv submission: January 21, 2026
- arXiv: 2601.14598
- Reviewed arXiv version: v2
- Sources reviewed: 2026-11-06
- Added to the catalog: 2026-11-06
- Original paper: <https://arxiv.org/abs/2601.14598>
- PDF: <https://arxiv.org/pdf/2601.14598>
- BibTeX: <https://awesomerea.org/papers/citations/paper-helios.bib>
- Catalog page: <https://awesomerea.org/papers/paper-helios/>
- Record JSON: <https://awesomerea.org/papers/paper-helios/record.json>
- Chinese version: [简体中文](https://awesomerea.org/zh/papers/paper-helios/index.md)

## Scope and limitations

Evaluation covers HumanEval-Decompile on x86_64 with Gemini 2.0 and GPT-4.1 Mini, plus six architectures from x86, ARM, and MIPS.

## Why it fits this collection

It addresses binary decompilation for reverse engineering by using program structure to help generate recompilable, functionally correct code.

## Publication and review notes

The date and venue come from the arXiv record.

## Primary sources

1. <https://arxiv.org/abs/2601.14598>

## 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.