Conditional Optimal Bridge for Riemannian Activation Steering
Cobras introduces Conditional Optimal Bridge for Riemannian Activation Steering, a method to control large language models during inference by steering their internal activations. It frames steering as finding a path between desired and undesired activation distributions using a Schrödinger Bridge on the residual-stream hypersphere. This provides a principled derivation of the log-density-ratio objective and creates query-adaptive steering directions that improve performance across helpfulness, truthfulness, and detoxification tasks while preserving out-of-distribution accuracy.