发现大模型用一维方向编码抽象程度，可无训练调控文本风格
Exploring Concreteness Through a Figurative Lens
- 分析四类大模型隐藏层，发现早期层区分字面与隐喻用法
- 中后期层将抽象度压缩成统一方向，跨模型一致
- 该方向可直接用于隐喻识别与生成风格控制
静态具象性评分在NLP中广泛应用，但词语的具象性会随语境变化，尤其在隐喻等修辞中，常见具体名词可能获得抽象含义。尽管这种变化在上下文中明显，但大模型如何内部理解具象性仍不清晰。本文对四种模型家族的隐藏表示进行分层与几何分析，研究模型如何区分同一名词的字面与隐喻用法，以及具象性在表示空间中的组织方式。结果表明，大模型在早期层即分离字面与隐喻使用，中至晚期层将具象性压缩为一条跨模型一致的一维方向。最后，我们证明该几何结构具有实际应用价值：单个具象性方向即可支持高效的隐喻语言分类，并实现无需训练的生成风格调控，使输出更字面化或更隐喻化。
原文摘要 · Abstract (English)
Static concreteness ratings are widely used in NLP, yet a word's concreteness can shift with context, especially in figurative language such as metaphor, where common concrete nouns can take abstract interpretations. While such shifts are evident from context, it remains unclear how LLMs understand concreteness internally. We conduct a layer-wise and geometric analysis of LLM hidden representations across four model families, examining how models distinguish literal vs figurative uses of the same noun and how concreteness is organized in representation space. We find that LLMs separate literal and figurative usage in early layers, and that mid-to-late layers compress concreteness into a one-dimensional direction that is consistent across models. Finally, we show that this geometric structure is practically useful: a single concreteness direction supports efficient figurative-language classification and enables training-free steering of generation toward more literal or more figurative rewrites.
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。