让大模型学会像人一样推理隐含意义，提升理解力
Understand the Implication: Learning to Think for Pragmatic Understanding
- 用人类思考过程训练模型，而非仅依赖标注标签
- 准确率提升11.12%，跨模型家族表现稳定
- 可迁移到预设、指示词等未见任务，提升16.10%
语用学是理解字面之外含义的关键能力，对社交认知与交流至关重要。尽管大语言模型已在语用理解上被评估，但其性能提升仍缺乏深入探索。现有方法依赖标注标签，忽视人类推断隐含意义的自然推理过程。为此，我们构建了全新语用数据集ImpliedMeaningPreference，包含正确与错误解释对应的显式推理（思想）。通过偏好优化与监督微调，我们证明基于思想的学习能显著提升大模型的语用理解能力，在多个模型家族中平均准确率提高11.12%。进一步的迁移学习研究显示，该方法在未训练过的语用任务（如预设、指示词）上相比标签训练模型提升16.10%。
原文摘要 · Abstract (English)
Pragmatics, the ability to infer meaning beyond literal interpretation, is crucial for social cognition and communication. While LLMs have been benchmarked for their pragmatic understanding, improving their performance remains underexplored. Existing methods rely on annotated labels but overlook the reasoning process humans naturally use to interpret implicit meaning. To bridge this gap, we introduce a novel pragmatic dataset, ImpliedMeaningPreference, that includes explicit reasoning (thoughts) for both correct and incorrect interpretations. Through preference-tuning and supervised fine-tuning, we demonstrate that thought-based learning significantly enhances LLMs' pragmatic understanding, improving accuracy by 11.12% across model families. We further discuss a transfer-learning study where we evaluate the performance of thought-based training for the other tasks of pragmatics (presupposition, deixis) that are not seen during the training time and observe an improvement of 16.10% compared to label-trained models.
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。