Text Has Curvature
Grover, Zeng, Xia, Faloutsos, Gordon · cs.LG · 2026-09-05 · 原文
Does natural language text have an intrinsic curvature? Language is increasingly modeled in curved geometries - hyperbolic spaces for hierarchy, mixed-curvature manifolds for compositional structure - yet a basic scientific question remains unresolved: what does curvature mean for text itself, in a way that is native to language rather than an artifact of the embedding space we choose? We argue that text does indeed have curvature, and show how to detect it, define it, and use it. To this end, we propose Texture, a text-native, word-level discrete curvature signal, and make three contributions. (a) Existence: We provide empirical and theoretical certificates that semantic inference in natural corpora is non-flat. (b) Definition: We define Texture as a signed two-axis curvature of the word-in-context belief field - the differential of reconciliation between prefix and suffix - measuring, via a debiased Schrödinger transport divergence, whether adding context from one side contracts the semantic effect of context from the other side (focus, positive) or expands it into competing continuations (fan-out, negative). (c) Utility: Texture is actionable: it serves as a general-purpose meas
讲义
讲义·推断 依据「原文」自动生成的结构化摘要（推断），非原文表述；以原文为准。
1. 人话版
Does natural language text have an intrinsic curvature?
Language is increasingly modeled in curved geometries - hyperbolic spaces for hierarchy, mixed-curvature manifolds for compositional structure - yet a basic scientific question remains unresolved: what does curvature mean for text itself, in a way that is native to language rather than an artifact of the embedding space we choose?
2. 领域脉络
本文类目：cs.LG，属于其所在研究脉络的最新进展。
3. 机制拆解
We argue that text does indeed have curvature, and show how to detect it, define it, and use it.
(a) Existence: We provide empirical and theoretical certificates that semantic inference in natural corpora is non-flat.
4. 证据与数字
摘要未给出量化结果——留意原文的实验与数据。
5. 反例与边界
To this end, we propose Texture, a text-native, word-level discrete curvature signal, and make three contributions.
6. 跨领域连接与意外收获
思考本文机制能否迁移到你正在跟进的问题。
7. 可复用方法
把本文机制与你手头项目对照，找一个两周内能验证的最小实验。
8. 术语表
精读时把不熟的术语记入此处，作为下次回忆的锚点。