Geometric Evolution Maps: Extracting Stable Concept Probes from Transformer Residual Streams 事件

Geometric Evolution Maps: Extracting Stable Concept Probes from Transformer Residual Streams arXiv:2605.25848v1 Announce Type: cross Abstract: Concept probes extracted from transformer residual streams are only as reliable as the layer from which they are extracted. The common practice of probing at a fixed late layer or at the peak of a separation score function ignores a fundamental structural feature: concept representations undergo substantial directional rotation during their assembly phas