AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

The Concrete-Arbitrary Gap: Kinship Reasoning in LLMs Is Not Indifferent to Presentation

arXiv · AI, language, vision and robotics · article · Sep 30, 2026 · UTC

We test whether large language models solve formally matched kinship problems equally well when relations are expressed in familiar vocabulary or by explicitly defined nonce predicates. Across 500 paired graphs, concrete accuracy exceeds arbitrary accuracy by 35.6 percentage points in local Qwen3.8-27B, 26.6 in Gemma 4 26B-A4B, 12.0 in Gemma 4 31B, and 5.4 in Qwen3.8-Max. All four paired gaps are statistically resolved. Reasoning budgets and prompt-language interventions can substantially reduce the difference, showing that it is modifiable rather than a fixed incapacity. The minimal conclusio

Read original source ↗ Open in workspace

recordType
paper
region
Global

Evidence & attribution

First collected: 2026-10-01T15:32:10.694Z. This is not the publication date.