Table of Content
First, we search the top security, software engineering, AI and programming language avenues for papers published between 2022 and 2024 using keywords such as "binary similarity" and "binary search".
The avenues include:
Security: S&P, Usenix Security, CCS, NDSS, TIFS, TDSC
SE: ICSE, ASE, FSE, ISSTA, TSE, TOSEM
AI: ICML, AAAI, IJCAI, NeurIPS, ACL, ICLR
PL: PLDI, POPL, OOPSLA, ECOOP
This process yeilds 28 papers, after inspection, we filtered 13 unrelated papers among them.
The 13 out-of-scope papers (Enhancement papers, non-AI papers, dynamic papers, etc.):
NDSS
2025
BinEnhance: An Enhancement Framework Based on External Environment Semantics for Binary Code Search
TSE
2022
Enhancing DNN-Based Binary Code Function Search With Low-Cost Equivalence Checking
ICSE
2024
BinAug: Enhancing Binary Similarity Analysis with Low-Cost Input Repairing
Cross-Inlining Binary Function Similarity Detection
ISSTA
2023
Improving Binary Code Similarity Transformer Models by Semantics-Driven Instruction Deemphasis
2024
CEBin: A Cost-Effective Framework for Large-Scale Binary Code Similarity Detection
TOSEM
2023
Towards Practical Binary Code Similarity Detection: Vulnerability Verification via Patch Semantic Analysis
Asteria-Pro: Enhancing Deep Learning-based Binary Code Similarity Detection by Incorporating Domain Knowledge
2024
ARCTURUS: Full Coverage Binary Similarity Analysis with Reachability-guided Emulation
NeurIPs
2024
Source Code Foundation Models are Transferable Binary Analysis Knowledge Bases
ECOOP
2024
Java Bytecode Normalization for Code Similarity Analysis
We recursively examined the references of the remaining 15 papers to find more related papers. Finally we get a total of 34 papers:
We further selected representative BFSD tools for evaluation. First, we excluded ten papers with unavailable tools. Next, we identified six representative tools published before 2023, based on citation count, GitHub stars, and and their use as baselines in prior work. Finally, we included two SOTA tools from 2023 and 2024 based on their GitHub stars.