← 研踪 PaperTrailarXiv:2611.23185 · v1
Polarity Composition Hypothesis: Causal Evidence from Negation in NLI
极性组合假说:来自自然语言推理中否定的因果证据
首次提交 2026-11-18 02:30
版本更新 2026-11-18 02:30
雷达观测 2026-11-18 12:34
摘要
We investigate how language models process negation in natural language inference (NLI). We propose the \textit{Polarity Composition Hypothesis}, a high-level causal model of this computation. Behavioral evaluations show that language-model predictions largely conform to the proposed hypothesis. Using Distributed Alignment Search, we identify causally effective subspaces aligned with the variables in the polarity-composition hypothesis. These representations transfer across negation forms and partially to a subset of MNLI. We release the dataset and code.
作者备注:Accepted to EACL 2027 (main conference)
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-18 02:30 · 最新观测版本
Polarity Composition Hypothesis: Causal Evidence from Negation in NLI
We investigate how language models process negation in natural language inference (NLI). We propose the \textit{Polarity Composition Hypothesis}, a high-level causal model of this computation. Behavioral evaluations show that language-model predictions largely conform to the proposed hypothesis. Using Distributed Alignment Search, we identify causally effective subspaces aligned with the variables in the polarity-composition hypothesis. These representations transfer across negation forms and partially to a subset of MNLI. We release the dataset and code.
查看此版本解读 →查看此版本原文 ↗