Composition Collapse: Stable Factual Knowledge Does Not Imply Compositional Reasoning 事件

Composition Collapse: Stable Factual Knowledge Does Not Imply Compositional Reasoning arXiv:2605.26789v1 Announce Type: new Abstract: Post-training is routinely evaluated through aggregate benchmark scores that treat multi-hop reasoning as a single capability -- as if a model that answers more questions correctly must be better at assembling facts. We show that this assumption can be misleading: recipes with statistically indistinguishable atomic knowledge produce composition behaviour separate