LLM4Decompile/sk2decompile/evaluation
BaiRiDreamer 239cba2673 feat(sk2decompile): add BringUpBench evaluation pipeline and results
Integrate BringUpBench evaluation into sk2decompile/evaluation/bringupbench/,
corresponding to Section A.6 of the paper (arXiv:2509.22114).

BringUpBench is a benchmark suite of 90 self-contained C programs (505 functions,
O0-O3). SK2Decompile achieves 42.3% compilation rate and 27.0% re-executability
rate, compared to IDA Pro's 23.6% / 21.7%.

Contents:
- scripts/: 5-step reproduction pipeline (compile, decompile, map, infer, eval)
- data/func_maps/: pre-built function-level mappings (source <-> pseudo <-> asm)
- data/infer_results/: SK2Decompile inference outputs for all opt levels
- reports/: per-opt-level evaluation result summaries (Markdown)
- config.env: template environment configuration
- README.md: comprehensive documentation with reproduction guide

Also updated sk2decompile/README.md to reference BringUpBench evaluation.
2026-02-12 00:02:25 +08:00
..
bringupbench feat(sk2decompile): add BringUpBench evaluation pipeline and results 2026-02-12 00:02:25 +08:00
data sk2decompile 2025-10-08 18:15:57 +08:00
metrics sk2decompile 2025-10-08 18:15:57 +08:00
evaluate_exe.py sk2decompile 2025-10-08 18:15:57 +08:00
evaluate_r2i.py sk2decompile 2025-10-08 18:15:57 +08:00
gpt_judge.py sk2decompile 2025-10-08 18:15:57 +08:00
inf_type.py sk2decompile 2025-10-08 18:15:57 +08:00
llm_server.py Add files via upload 2025-10-16 23:29:55 +08:00
normalize_pseudo.py Add files via upload 2025-10-16 23:29:55 +08:00
sk2decompile_inf.py Add files via upload 2025-10-16 23:29:55 +08:00
stack.yaml sk2decompile 2025-10-08 18:15:57 +08:00
template.txt sk2decompile 2025-10-08 18:15:57 +08:00