跳到正文
arXiv cs.CL· Huixue Han, Jiaying Liu·· 10 小时前AI 评分26

研究:事实层应提供计算出的交互关系而非静态属性——以中国干支系统为例

Compute the Interactions, Not the Attributes: What a Fact Layer Must Supply in the Chinese Sexagenary System

AI 导读

针对干支(BaZi)排盘这类规则明确领域,研究测量了事实层应提供哪些事实:682 个条目覆盖十二类事实,另有 160 个行运与本命碰撞项。十一个开放权重模型直接作答时,干支日正确率均不超 4%、碰撞项不超 1%;完整关系列表虽含答案,多数模型仍接近零。

正文 · 原文

View PDF HTML (experimental)

Abstract:Language-model systems in rule-governed domains often pair the model with a deterministic fact layer, but which facts that layer should supply is rarely measured. We measure it in the Chinese sexagenary system underlying BaZi chart analysis, where every fact has one table-checkable value: 682 items over twelve fact categories and 160 transit--natal collision items, posed closed-book, with the reference table, and with computed facts at increasing levels. In eleven open-weight models answering directly, errors concentrate where attributes must be combined: none gets more than 4\% of sexagenary days or 1\% of collisions right. Static attributes have no consistent effect, and the complete relation list leaves most direct-answering models near zero although it contains the answer. A query-filtered layer (exploratory) raises Qwen3-8B and Qwen3-32B to 78--92\% and four further models to 96--100\%. Three proprietary reasoning models compute 68--95\% of collisions unaided. On chess positions, under predictions locked in advance, attributes again do not substitute for relations and listed distractors are misattributed. The fact layer must supply computed interactions; how they are presented depends on the model and the domain.
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2611.13597 [cs.CL]
  (or arXiv:2611.13597v1 [cs.CL] for this version)
  https://doi.org/10.48550/arXiv.2611.13597

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Huixue Han [view email]
[v1] Tue, 10 Nov 2026 05:15:58 UTC (352 KB)

来源:arXiv cs.CL · arxiv.org