LibEvoBench benchmark shows LLMs are version-oblivious on evolving APIs, with documentation helping but version specification not.
Codesync: Synchronizing large language models with dynamic code evolution at scale
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.SE 2years
2026 2representative citing papers
A closed-loop multi-level evolving memory that stores API-, cross-API-, and task-level usage guidelines from execution feedback substantially raises LLM pass rates on private-library code generation.
citing papers explorer
-
LibEvoBench: Probing Temporal Knowledge Stratification in Code Generation Models
LibEvoBench benchmark shows LLMs are version-oblivious on evolving APIs, with documentation helping but version specification not.
-
Learning from Execution: Self-Evolving Memory for Private-Library Code Generation
A closed-loop multi-level evolving memory that stores API-, cross-API-, and task-level usage guidelines from execution feedback substantially raises LLM pass rates on private-library code generation.