Constrained decoding guarantees that Large Language Models (LLMs) produce outputs conforming to predefined formats, but is often assumed to inherently degrad…
机构:Nokia Bell Labs
来源:arXiv 2610.35328 | AI4Papers 论文推荐平台