Math benchmarks may miss the reasoning strategies models actually reuse
Math-capable language models appear to organize computation by reusable reasoning approach, not by topic, so topic-balanced benchmarks can still test an uneven mix of strategies. The study replays each model’s solution trajectory, clusters activation-importance patterns, and finds approach-level coherence in 77–82% of real clusters versus 6–11% in controls. Changing the requested approach shifts the cluster in seven of eight model conditions, while paraphrasing usually does not.