让大模型学会合理分配任务，高效准确回答问题。
Optimal Query Allocation in Extractive QA with LLMs: A Learning-to-Defer Framework with Theoretical Guarantees
- 设计学习延迟机制，智能分配查询给专业模型
- 在多个数据集上提升答案可靠性，降低计算开销
- 有理论保证的优化策略，适合资源受限场景
大型语言模型在生成任务中表现优异，但在结构化文本选择（尤其是抽取式问答）中效率低下。这一问题在资源受限环境下尤为突出，因部署多个专用模型不现实。本文提出一种学习延迟框架，将查询分配给专业专家模型，在确保高置信度预测的同时优化计算效率。该方法结合了严谨的分配策略与最优延迟的理论保证，平衡性能与成本。在 SQuADv1、SQuADv2 以及 TriviaQA 数据集上的实证评估表明，该方法显著提升了答案可靠性并大幅减少计算开销，适用于可扩展、高效的抽取式问答系统部署。
原文摘要 · Abstract (English)
Large Language Models excel in generative tasks but exhibit inefficiencies in structured text selection, particularly in extractive question answering. This challenge is magnified in resource-constrained environments, where deploying multiple specialized models for different tasks is impractical. We propose a Learning-to-Defer framework that allocates queries to specialized experts, ensuring high-confidence predictions while optimizing computational efficiency. Our approach integrates a principled allocation strategy with theoretical guarantees on optimal deferral that balances performance and cost. Empirical evaluations on SQuADv1, SQuADv2, and TriviaQA demonstrate that our method enhances answer reliability while significantly reducing computational overhead, making it well-suited for scalable and efficient EQA deployment.
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。