Steering by Influence: Curvature Aware Data Weighting for Activation Steering

Excerpt

arXiv:2610.06383v1 Announce Type: cross Abstract: Inference-time steering offers cheap, fine-grained control over a language model's outputs by estimating a concept's representation in activation space and shifting activations towards it. Existing methods build these representations from activation averages over contrastive datasets. These averages incorporate unrelated concepts and noise, and are dominated by a few tokens, meaning the activation transport encodes token-level rather than themati