让大模型生成更准：基于任务结构的最优响应合成
Task-Awareness Improves LLM Generations and Uncertainty
- 在任务相关隐空间中直接建模输出结构，而非仅在语言空间解码
- 贝叶斯最优响应在多任务上均优于束搜索等标准方法
- 通过贝叶斯风险量化不确定性，与输出质量高度对齐
在许多大模型应用中，自然语言输出通常具有潜在结构，如离散标签、数值或图结构。然而，现有解码与不确定性估计方法仅在语言空间操作，忽视了结构信息。本文提出在任务相关的隐空间中直接建模输出，通过引入相异度度量，计算贝叶斯最优响应——并非从采样结果中选择，而是通过隐空间组合生成。该方法在多个任务上均优于束搜索等标准解码方式。此外，基于诱导贝叶斯风险的不确定性量化能捕捉隐结构变化，提升与输出质量及正确性的对齐。本决策理论框架适用于任何具备隐响应结构的问题，支持可靠的任务感知大模型预测。
原文摘要 · Abstract (English)
In many applications of LLMs, natural language responses often have an underlying structure such as representing discrete labels, numerical values, or graphs. Yet, existing decoding and uncertainty estimation methods operate only in language space and largely disregard structural information. We address this by modeling LLM outputs directly in a task-dependent latent structure. By equipping this structure with a dissimilarity measure, we can compute Bayes-optimal responses. These are not selected from sampled generations but are newly synthesized by combining individual responses in the latent space. Across different tasks, Bayes-optimal responses consistently outperform standard decoding methods like beam search. Moreover, quantifying uncertainty via the induced Bayesian risk captures variations in terms of the latent structure and improves alignment with output quality and correctness. Our decision-theoretic framework is applicable to any problem that admits a latent response structure and enables reliable task-aware LLM predictions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。