CodeSpecBench shows LLMs achieve at most 20.2% pass rate on repository-level executable behavioral specification generation, revealing that strong code generation does not imply deep semantic understanding.
An empirical evaluation of pre-trained large language models for repairing declarative formal specifications
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.SE 1years
2026 1verdicts
ACCEPT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
CodeSpecBench: Benchmarking LLMs for Executable Behavioral Specification Generation
CodeSpecBench shows LLMs achieve at most 20.2% pass rate on repository-level executable behavioral specification generation, revealing that strong code generation does not imply deep semantic understanding.