AI RESEARCH
Schema Key Wording as an Instruction Channel in Structured Generation under Constrained Decoding
arXiv CS.AI
•
ArXi:2604.14862v1 Announce Type: cross Constrained decoding has been widely adopted for structured generation with large language models (LLMs), ensuring that outputs satisfy predefined formats such as JSON and XML. However, existing approaches largely treat schemas as purely structural constraints and overlook the possibility that their linguistic formulation may affect model behavior.