First, we formalize the problem definition.Weintroducetheconceptof"Idon'tknow (idk) responses" and in this context, honesty necessitates that an aligned LLM provides idk responses for unknown questions and correct responses for known questions.

These authors contributed equally and are ordered alphabetically.

SelfCodeAlign employs the same base model for inference throughout the data generation process. It first extracts diverse coding concepts from high-quality seed snippets to generate new tasks.