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

• URL: http://arxiv.org/abs/2608.27981v1
• Date: Fri, 28 Aug 2026 06:49:21 GMT
• Title: CHISEL-ing Back Source Code with AI-enabled Iterative Recovery
• Abstract summary: Decompilation aims to recover high-level, compilable, and semantically equivalent code from binaries.<n>Traditional decompilers produce pseudo-C that is difficult to read and does not compile, while the recent LLM-assisted approaches generate readable, but semantically incorrect code.<n>We present CHISEL, a test suite-free framework to iteratively recover source code from Ghidra-derived pseudo-C.
• Score: 7.946261291068861
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Decompilation aims to recover high-level, compilable, and semantically equivalent code from binaries. Traditional decompilers produce pseudo-C that is difficult to read and does not compile, while the recent LLM-assisted approaches generate readable, but semantically incorrect code. LLM-aided iterative recovery is an emerging branch of research, but prior works rely on supplied test suites for semantic recovery. In this work, we present CHISEL, a test suite-free framework to iteratively recover source code from Ghidra-derived pseudo-C. CHISEL uses simple yet effective feedback from a compiler (static analysis) and a coverage-guided fuzzer (differential analysis), augmented by rich observables for grounded divergence detection and feedback, cross-iteration divergence memory, and best candidate retention. We systematically evaluate CHISEL for compilation and semantic recovery, feedback oracle soundness, and iteration overhead on 120 ExeBench functions compiled for the x86-64 architecture, across four optimizations (O0-O3), in both stripped and unstripped variants, using the open-weight Gemma4:31b LLM. CHISEL, with all recommended features, achieves an average of 96.1% re-compilability and 79.8% re-executability rates at an average of 2.1 iterations. Significantly, CHISEL recovers 26% of first-generation execution errors. At the same time, CHISEL feedback oracle falsely accepts only 9.4% candidates. Lastly, CHISEL performs significantly better than two recent prior work on LLM-assisted decompilation.

Related papers

• When LLM Decompilers Recompile More and Preserve Less [35.2873296518414]
  We show that a function may and pass every shipped test yet diverge on other legitimate inputs.<n>We propose De-Diverge, a behavioral comparison oracle not relying on fixed or hand-crafted tests.
  arXiv  Detail & Related papers  (2026-09-04T17:16:18Z)
• STILL: Recovering Lowered STL Semantics for LLM-assisted C++ Decompilation [51.12395225250876]
  We present STILL, a structured semantic interface that predicts function-level STL container semantics from stripped control-flow graphs.<n>On StlBench, STILL predicts common container-level STL semantics, with the strongest cross-dataset results for stable string and vector slices.<n>On stripped HumanEval decompilation, these hints enable DeepSeek-chat refinement to reach 28.4% executability, compared with 17.4% for no-hint refinement.
  arXiv  Detail & Related papers  (2026-08-26T21:20:54Z)
• Can Large Language Models Recover Semantic Optimization Opportunities That Compilers Miss? [5.989315760622425]
  Large language models (LLMs) can recover enabling semantics from heterogeneous C/C++ context.<n>We introduce SeGaBench, an executable benchmark containing 100 synthetic and 20 source-backed cases.<n>The strongest model produces correct artifacts in 94.8% of responses, achieves at least 1.05x speedup in 83.3%, and obtains a performance success on 93.3% of cases.
  arXiv  Detail & Related papers  (2026-08-04T17:47:25Z)
• CODEFUSE-DEBENCH: An Empirical Study on Readability, Recompilability, and Functionality [14.354022325641088]
  We propose a reusability-driven evaluation paradigm that measures decompiler quality along three dimensions: readability, recompilability, and functionality.<n>We present DEBENCH, the first automated framework for multidimensional decompilation evaluation.
  arXiv  Detail & Related papers  (2026-05-28T07:17:53Z)
• Constraint-Guided Multi-Agent Decompilation for Executable Binary Recovery [47.704311990064554]
  Decompilation is essential for security analysis, malware reverse engineering, and legacy software maintenance.<n>We present a multi-agent framework that transforms decompiled code into re-executable source through Multi-level Constraint-Guided Decompilation (MCGD)<n>Our framework achieves 84-97% re-executability, improving baseline decompiler output by 28-89 percentage points.
  arXiv  Detail & Related papers  (2026-04-27T01:28:11Z)
• Resource-Efficient Iterative LLM-Based NAS with Feedback Memory [49.44875022114861]
  Neural Architecture Search (NAS) automates network design, but conventional methods demand substantial computational resources.<n>We propose a closed-loop pipeline leveraging large language models (LLMs) to iteratively generate, evaluate, and refine convolutional neural network architectures.
  arXiv  Detail & Related papers  (2026-03-12T16:00:22Z)
• Context-Guided Decompilation: A Step Towards Re-executability [50.71992919223209]
  Binary decompilation plays an important role in software security analysis, reverse engineering and malware understanding.<n>Recent advances in large language models (LLMs) have enabled neural decompilation, but the generated code is typically only semantically plausible.<n>We propose ICL4Decomp, a hybrid decompilation framework that leverages in-context learning (ICL) to guide LLMs toward generating re-executable source code.
  arXiv  Detail & Related papers  (2025-11-03T17:21:39Z)
• QiMeng-NeuComBack: Self-Evolving Translation from IR to Assembly Code [52.66657751895655]
  Large Language Models (LLMs) offer a compelling new paradigm: Neural Compilation.<n>This paper introduces NeuComBack, a novel benchmark dataset specifically designed for IR-to-assembly compilation.<n>We propose a self-evolving prompt optimization method that enables LLMs to evolve their internal prompt strategies.
  arXiv  Detail & Related papers  (2025-11-03T03:20:26Z)
• D-LiFT: Improving LLM-based Decompiler Backend via Code Quality-driven Fine-tuning [49.16469288280772]
  Decompilers reconstruct human-readable source code from binaries.<n>Despite recent advances, their outputs often suffer from syntactic and semantic errors and remain difficult to read.<n>With the advent of large language models (LLMs), researchers began to explore the potential of LLMs to refine decompiler output.<n>We present D-LIFT, an enhanced decompiler-LLM pipeline with fine-tuned reinforcement learning.
  arXiv  Detail & Related papers  (2025-06-11T19:09:08Z)
• ALGO: Synthesizing Algorithmic Programs with LLM-Generated Oracle Verifiers [60.6418431624873]
  Large language models (LLMs) excel at implementing code from functionality descriptions but struggle with algorithmic problems. We propose ALGO, a framework that synthesizes Algorithmic programs with LLM-Generated Oracles to guide the generation and verify their correctness. Experiments show that when equipped with ALGO, we achieve an 8x better one-submission pass rate over the Codex model and a 2.6x better one-submission pass rate over CodeT.
  arXiv  Detail & Related papers  (2023-05-24T00:10:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.