# CHISEL-ing Back Source Code with AI-enabled Iterative Recovery

> CHISEL refines Ghidra pseudocode using compiler diagnostics and type-aware differential fuzzing against the original function. It retains counterexamples and candidate history; existing I/O pairs are held out for evaluation rather than supplied to the recovery loop.

- Authors: Varun Kohli; N Raghava; Biplab Sikdar; Dinil Mon Divakaran
- Venue: SURE @ ACM CCS 2026
- Status: Conference accepted
- Publication / posting date: 2026 (exact date not verified)
- Date source: Acceptance is reported in the author arXiv record and publications page; proceedings publication is not verified.
- First arXiv submission: August 28, 2026
- arXiv: 2608.27981
- Reviewed arXiv version: v1
- Sources reviewed: 2026-10-20
- Added to the catalog: 2026-10-20
- Original paper: <https://arxiv.org/abs/2608.27981>
- PDF: <https://arxiv.org/pdf/2608.27981v1>
- Code: <https://github.com/VarunKohli18/chisel>
- BibTeX: <https://awesomerea.org/papers/citations/paper-chisel-iterative-recovery.bib>
- Catalog page: <https://awesomerea.org/papers/paper-chisel-iterative-recovery/>
- Record JSON: <https://awesomerea.org/papers/paper-chisel-iterative-recovery/record.json>
- Chinese version: [简体中文](https://awesomerea.org/zh/papers/paper-chisel-iterative-recovery/index.md)

## Scope and limitations

Evaluation covers 120 selected ExeBench functions, x86-64 O0–O3, stripped/unstripped variants and a fixed Gemma4:31b model. The feedback oracle still falsely accepts candidates; finite fuzzing and held-out tests do not establish equivalence. Signature recovery and possible benchmark exposure remain limitations.

## Why it fits this collection

Adds test-suite-free in-loop refinement while explicitly separating oracle acceptance from verified behavior.

## Publication and review notes

Author sources report acceptance to SURE 2026; the manuscript DOI remains a placeholder, so no proceedings publication date or DOI is assigned. The paper says selection requires at least 200 tests, while the pinned README says at least 100 held-out I/O pairs; exact evaluated artifact alignment remains unresolved. The article is CC BY 4.0, but the inspected root/README did not establish a code/data license.

## Primary sources

1. <https://arxiv.org/abs/2608.27981v1>
2. <https://arxiv.org/html/2608.27981v1>
3. <https://www.dinil.net/publications>
4. <https://github.com/VarunKohli18/chisel/blob/855acb07d0ee69b759d91895e8b5a846291db5a5/README.md>
5. <https://api.github.com/repos/VarunKohli18/chisel/git/trees/ece4cc2b45641e4aa968f2343f165c5ca242d5da>

## 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.