Indirect Function Call Target Identification within Software
By employing explicit data dependency analysis and a graph neural network to identify similar functions, the method enhances the precision of indirect function call target identification, addressing the inaccuracies in current methods and improving program security analysis.
JP7893575B2Active Publication Date: 2026-07-22INTERNATIONAL BUSINESS MACHINE CORPORATION
View PDF 3 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- INTERNATIONAL BUSINESS MACHINE CORPORATION
- Filing Date
- 2022-11-03
- Publication Date
- 2026-07-22
AI Technical Summary
Technical Problem
Current methods for identifying indirect function call targets in software are not precise enough for large programs, leading to high false-positive and false-negative rates, which hinders effective program analysis such as fuzz testing and vulnerability detection.
Method used
A method using explicit data dependency analysis and a graph neural network to generate feature embeddings, identifying similar functions, and expanding the ground truth dataset of indirect function call targets to improve accuracy.
Benefits of technology
Reduces false-positive and false-negative rates, enabling precise identification of indirect function call targets for improved program security analysis.
✦ Generated by Eureka AI based on patent content.
Smart Images

Figure 0007893575000001 
Figure 0007893575000002 
Figure 0007893575000003
Abstract
Indirect function call target identification within the software is provided. A set of explicit data flows that pass function addresses between software modules of the program is determined using explicit data dependency analysis. A set of indirect function call targets is generated from results of the explicit data dependency analysis and dynamic execution analysis of the program. The set of indirect function call targets is expanded by identifying similar target functions based on feature embeddings generated by the graph neural network.
Need to check novelty before this filing date? Find Prior Art