The AutoDecompiler paper presents feedback-driven, multi-turn refinement of decompiled code using compilation, execution and input/output tests. Its useful idea is to treat a generated program as a hypothesis that can fail observable checks. Read the reported benchmark conditions carefully: compilation success alone does not establish the same behavior as the original binary.
Research question
Compare the authors' iterative feedback setup with a one-shot baseline under the same inputs and model size. Note which failures can be detected by the provided tests and which behaviors remain unobserved.
Availability
The current arXiv title is Binary Decompilation LLM with Feedback-Driven Multi-Turn Refinement. The record describes AutoDecompiler, but code, dataset and model availability are Unknown unless an author source explicitly establishes them.
Requirements & evidence
- Resource type
- Paper
- Category
- Unknown
- Host platforms
- Unknown
- Targets
- Unknown
- Level
- Unknown
- Author(s)
- Peipei Liu, Jian Sun, Mingzhe Xing, Yicheng Zeng, Zhaoteng Yan, Lixiao Zhang, Li Chen, Dan Li
- Year
- 2026
- Venue
- arXiv preprint
- Current version
- v2
- Version date
- 2026-10-03
- Topic
- AutoDecompiler iterative decompilation with compile and execution feedback
- Source checked
- 2026-11-03T10:44:20.012388+00:00
Unknown fields have not been established by an author or publisher source.
Primary sources
Source descriptions establish documented scope. They do not establish that every workflow has been reproduced by AI Craft Pad.