Research map / Graph learning / Graph self-supervised learning
Order Matters: Semantic-Aware Neural Networks for Binary Code Similarity Detection
Zeping Yu, Rui Cao, Qiyi Tang, Sen Nie, Junzhou Huang, Shi Wu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| full model (BERT4+MPNN+Resnet11) vs Gemini | binary code similarity · Task1-O2 | MRR10: 0.6069 -> 0.7922 | Improved |
| full model vs Gemini | binary code similarity · Task2-ARM | accuracy: 79.89 -> 88.41 | Improved |
| BERT pre-trained block features instead of manual features (ablation) | MPNN · Task1-O2 | MRR10: 0.6096 -> 0.7591 | Improved |
| add graph-level pre-training tasks BIG and GC (ablation) | BERT block embeddings · Task2-x86-64 | accuracy: 82.67 -> 83.74 | Improved |
| add CNN order-aware module on adjacency matrix (ablation) | BERT4+MPNN · Task1-O2 | MRR10: 0.7704 -> 0.7922 | Improved |
| randomly shuffle node order before CNN (ablation) | CNN3 · Task1-O2 | MRR10: 0.4142 -> 0.0362 | Hurt |
| deeper CNN (Resnet11 vs CNN3) (ablation) | order-aware CNN · Task1-O2 | MRR10: 0.4142 -> 0.4419 | Improved |
| combine MPNNws with Resnet11 (ablation) | MPNN without semantics · Task1-O2 | MRR10: 0.3361 -> 0.4457 | Improved |
Closest papers by meaning
- Efficient Code Analysis via Graph Representation Learning-Guided Large Language Models (ICML 2026 · similarity 0.87)
- Graph Matching Networks for Learning the Similarity of Graph Structured Objects (ICML 2019 · similarity 0.85)
- Transforming Generic Coder LLMs to Effective Binary Code Embedding Models for Similarity Detection (NeurIPS 2025 · similarity 0.85)
- GraphCodeBERT: Pre-training Code Representations with Data Flow (ICLR 2021 · similarity 0.85)
- CodeCMR: Cross-Modal Retrieval For Function-Level Binary Source Code Matching (NeurIPS 2020 · similarity 0.85)
- VQGraph: Rethinking Graph Representation Space for Bridging GNNs and MLPs (ICLR 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
