用语义图增强模型，反而让推理更差
When Does Meaning Backfire? Investigating the Role of AMRs in NLI
- 在微调中加入语义图，反而降低模型泛化能力
- 提示词中使用语义图，仅使GPT-4o略有提升
- 提升源于表面差异放大，非真正语义理解
自然语言推理（NLI）高度依赖对前提和假设语义内容的准确解析。本文研究以抽象语义表示（AMR）形式添加语义信息，是否有助于预训练语言模型在NLI任务中更好泛化。实验显示，在微调中引入AMR会阻碍模型泛化，而提示时使用AMR则使GPT-4o获得轻微提升。消融分析表明，该提升源于放大表面差异，而非促进语义推理。这种放大可能误导模型将语义一致的句子误判为非蕴含。
原文摘要 · Abstract (English)
Natural Language Inference (NLI) relies heavily on adequately parsing the semantic content of the premise and hypothesis. In this work, we investigate whether adding semantic information in the form of an Abstract Meaning Representation (AMR) helps pretrained language models better generalize in NLI. Our experiments integrating AMR into NLI in both fine-tuning and prompting settings show that the presence of AMR in fine-tuning hinders model generalization while prompting with AMR leads to slight gains in GPT-4o. However, an ablation study reveals that the improvement comes from amplifying surface-level differences rather than aiding semantic reasoning. This amplification can mislead models to predict non-entailment even when the core meaning is preserved.
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。