# A Memorization Floor for LLM Refinement of Decompiled Code

> Studies whether LLM refinement of decompiled C recovers identifiers from input evidence or prior knowledge. Within-item ablations and output-matched chance baselines test naming gains on textbook and newly authored functions, including a second-model replication.

- Authors: Muhammad Asjad
- Venue: arXiv
- Status: Preprint
- Publication / posting date: September 15, 2026
- Date source: arXiv v1 submission date; the artifact deposit date is separate.
- First arXiv submission: September 15, 2026
- arXiv: 2609.17236
- Reviewed arXiv version: v1
- DOI: 10.48550/arxiv.2609.17236
- Sources reviewed: 2026-10-20
- Added to the catalog: 2026-10-20
- Original paper: <https://arxiv.org/abs/2609.17236>
- PDF: <https://arxiv.org/pdf/2609.17236v1>
- Project: <https://zenodo.org/records/22121473>
- BibTeX: <https://awesomerea.org/papers/citations/paper-memorization-floor.bib>
- Catalog page: <https://awesomerea.org/papers/paper-memorization-floor/>
- Record JSON: <https://awesomerea.org/papers/paper-memorization-floor/record.json>
- Chinese version: [简体中文](https://awesomerea.org/zh/papers/paper-memorization-floor/index.md)

## Scope and limitations

The study uses 25 textbook and 20 author-written small functions, GCC/Ghidra on x86-64, and one prompt per model. Operations remain visible and the ablation also changes idiom-recognition cues, so its bounded null does not prove memorization or absence of input use. No registered hypothesis was confirmed; the correctness measure failed its discrimination gate; original T3 per-item artifacts were lost and re-executed, changing precision. Most findings use one refiner, and the later second-model extension is distinct from the initial preregistration.

## Why it fits this collection

Provides controls for interpreting apparently readable or well-named recovered code and cautions against treating naming scores as semantic recovery.

## Publication and review notes

Single-author arXiv v1; no formal venue was verified. DataCite confirms the arXiv DOI was registered on 2026-09-16 despite the abstract page’s pending label. The manuscript uses arXiv’s non-exclusive distribution license. The paper’s Zenodo concept DOI resolves to record 22121473, version 1.1.0, dated 2026-08-27, declaring CC BY 4.0 for a 9,639,907-byte replication ZIP.

## Primary sources

1. <https://arxiv.org/abs/2609.17236>
2. <https://arxiv.org/html/2609.17236v1>
3. <https://api.datacite.org/dois/10.48550/arxiv.2609.17236>
4. <https://doi.org/10.5281/zenodo.21968878>
5. <https://zenodo.org/records/22121473>
6. <https://zenodo.org/api/records/21968878>

## 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.