跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 28 天前AI 評分47

受約束解碼有害還是有益?重新探討其對大語言模型表現的影響

Constrained Decoding: Harmful or Beneficial? Revisiting Its Impact on LLM Performance

AI 導讀

研究發現,受約束解碼不會必然損害大語言模型的推理能力,其影響取決於約束類型。在四個模型系列的十一個開源模型上進行的實驗顯示,表面上的性能下降主要源於提示詞及輸出格式不一致,而非解碼器層面的遮罩;有效性約束可提升可靠性,推理約束在限制有益推理軌跡時亦可提高準確度。

正文

Published on Oct 26

Authors:

,

,

Abstract

Constrained decoding does not inherently harm reasoning; its effects depend on constraint type, with validity and reasoning constraints often improving reliability and accuracy by removing invalid paths.

Constrained decoding guarantees that Large Language Models (LLMs) produce outputs conforming to predefined formats, but is often assumed to inherently degrade reasoning. We examine this assumption through a taxonomy of three constraint types: format constraints, which enforce output structure; validity constraints, which mask invalid answer; and reasoning constraints, which restrict reasoning trajectories. We theoretically show that appropriate constraints can preserve or improve performance by eliminating invalid paths. Across four model families and eleven open-source models, causal ablations on diverse benchmarks using state-of-the-art constrained decoding strategies reveal that apparent performance degradation largely stems from non-equivalent prompting and output formats rather than from decoder-level masking. Meanwhile, validity constraints improve reliability and extractability by eliminating invalid answers, while reasoning constraints can improve reasoning and, consequently, accuracy when they enforce beneficial trajectories. These findings show that the impact of constrained decoding depends on the type of constraint and the output space it removes. Our code is available.

View arXiv page View PDF Add to collection

Get this paper in your agent:

hf papers read 2610.35328

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.35328 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.35328 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.35328 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co