SWI：大语言模型中的意图驱动对话

SWI: Speaking with Intent in Large Language Models

March 27, 2025

作者: Yuwei Yin, EunJeong Hwang, Giuseppe Carenini

cs.AI

摘要

意图，通常明确制定并规划，作为推理和问题解决的认知框架发挥作用。本文在大语言模型（LLMs）中引入了“有意图对话”（Speaking with Intent, SWI）的概念，其中明确生成的意图封装了模型的潜在意图，并提供高层规划以指导后续的分析与交流。通过模拟人类思维中深思熟虑且目标明确的思考过程，SWI被假设为能够增强LLMs的推理能力和生成质量。在数学推理基准上的大量实验一致表明，有意图对话相较于基线（即无明确意图的生成）具有显著优势。此外，SWI在触发答案提示方法如“思维链”（Chain-of-Thought）和“计划与解决”（Plan-and-Solve）之上表现更优，并与强方法ARR（分析、检索与推理）保持竞争力。同时，SWI在推理密集型问答（QA）和文本摘要基准上的有效性和泛化能力得到巩固，为基线生成带来了持续的改进。在文本摘要任务中，SWI生成的摘要展现出更高的准确性、简洁性和事实正确性，幻觉现象更少。进一步地，人类评估验证了SWI生成意图的连贯性、有效性和可解释性。这项概念验证研究为利用认知概念增强LLMs的推理能力开辟了一条新途径。

English

Intent, typically clearly formulated and planned, functions as a cognitive framework for reasoning and problem-solving. This paper introduces the concept of Speaking with Intent (SWI) in large language models (LLMs), where the explicitly generated intent encapsulates the model's underlying intention and provides high-level planning to guide subsequent analysis and communication. By emulating deliberate and purposeful thoughts in the human mind, SWI is hypothesized to enhance the reasoning capabilities and generation quality of LLMs. Extensive experiments on mathematical reasoning benchmarks consistently demonstrate the superiority of Speaking with Intent over Baseline (i.e., generation without explicit intent). Moreover, SWI outperforms answer-trigger prompting methods Chain-of-Thought and Plan-and-Solve and maintains competitive performance with the strong method ARR (Analyzing, Retrieving, and Reasoning). Additionally, the effectiveness and generalizability of SWI are solidified on reasoning-intensive question answering (QA) and text summarization benchmarks, where SWI brings consistent improvement to the Baseline generation. In text summarization, SWI-generated summaries exhibit greater accuracy, conciseness, and factual correctness, with fewer hallucinations. Furthermore, human evaluations verify the coherence, effectiveness, and interpretability of the intent produced by SWI. This proof-of-concept study creates a novel avenue for enhancing LLMs' reasoning abilities with cognitive notions.