Tool Use Reduces Depth-Induced Collapse in OOD Reasoning

תקציר מקורי באנגליתarXiv:2602.21061v2 Announce Type: replace Abstract: Many current paths to more advanced AI depend on the assumption that large language models (LLMs) can generalize learned relationships to solve complex, out-of-distribution (OOD) problems. However, this is not an easy quality to measure. For most real-world problems and benchmarks it suffices to exploit a few memorized subproblems or a small fraction of the available data to produce a correct solution. This is interpolation. Generalization requires the capacity to make use of all available data to solve problems without these shortcuts. To test this quality, we introduce a synthetic Boolean circuit reconstruction benchmark over $GF(2)$. We use an adversarial sampling oracle to block partial-information shortcuts, ensuring that each step r