SOURCE-LINKED INTELLIGENCE

Shared circuits predict whether LLMs generalize across formats in arithmetic reasoning

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

In many forms of reasoning, including arithmetic reasoning, generalizing across superficial changes in input format is effortless for humans: anyone who can solve 2+5 can also solve 'two plus five'. In contrast, LLMs are more brittle to surface variations of the prompts: for example, they solve numeric arithmetic problems almost perfectly but are substantially less accurate on verbal renditions of the same problems. Here, we ask whether generalization across formats can be predicted from the models' internals. Using attribution patching, we first independently localize the circuit that each mo

Read original source ↗ Open in workspace

recordType

paper

region

Global