Semantic Compression for Memory Retention in LLM Test Generation
Abstract
We introduce a memory retention mechanism based on semantic compression for LLM-driven test generation, achieving higher consistency and coverage by leveraging past information. We confirmed an average compression ratio of 16%, effectively utilizing both code and execution information.
Fields of Study & Subjects
Software Testing and Debugging Techniques
Software Engineering Research
VLSI and Analog Circuit Testing
Publication details
- Journal / source
- IEICE Transactions on Information and Systems
- Volume
- E109.D
- Issue
- 9
- Pages
- 1446–1450
- Publication date
- 12 February 2026
- Language
- EN