arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2604.18296cs.CL
收藏

通过隐喻视角探索具体性

Exploring Concreteness Through a Figurative Lens

University of Cincinnati(俄亥俄州立大学)

Saptarshi Ghosh, Tianyu Jiang

AI总结:

研究探讨LLM如何区分字面与隐喻用法,发现早期层分离具体性,中后期层压缩为一维方向,该结构可提升隐喻分类和生成控制。

摘要:

静态的具体性评分广泛应用于NLP,但词语的具体性会随上下文变化,尤其在隐喻等修辞中,常见具体名词可取抽象解释。尽管这种转变明显来自上下文,但LLM内部如何理解具体性仍不明确。我们对四个模型家族的隐藏表示进行了分层和几何分析,研究模型如何区分相同名词的字面与隐喻用法,以及具体性在表示空间中的组织方式。发现LLM在早期层分离字面与隐喻用法,中后期层将具体性压缩为一致的一维方向。最后,我们证明这种几何结构具有实用性:单一具体性方向支持高效隐喻分类,并能实现无需训练的生成控制,偏向更字面或更隐喻的重写。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑