AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

The Information Geometry of Softmax: Probing and Steering

Kiho Park, Todd Nief, Yo Joong Choe, Victor Veitch

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
steer in the dual space (dual steering) instead of adding the probe in the primal space (Euclidean steering) (ablation)Gemma-3-4B steering, English to French concept · C4 contexts, off-target preservation vs target concept probabilityoff-target distribution preservation (counterfactual mass, KL, rank difference)Improved
steer in the dual space instead of the primal space (ablation)MetaCLIP-2 on synthetic objects, yellow to green · synthetic object imagesoff-target distribution preservationImproved
steer in the dual space instead of the primal space (ablation)MetaCLIP-2 on COCO, carrot to broccoli · COCO imagesoff-target distribution preservationImproved
use dual mean difference instead of primal mean difference as probe direction for dual steering (ablation)dual steering · LLM and CLIP steeringoff-target preservationNo clear effect
Euclidean steering with primal mean difference probe (ablation)Euclidean steering · multiple concepts, KL of off-target distributionoff-target KL divergenceMixed
add regularization to the covariance in the Newton stepdual steering implementation · concentrated softmax contextsfeasibility of steering stepImproved
violate probe invariance assumption (probe entangled with off-target concepts)dual steering · finite-data probesoff-target preservation vs EuclideanNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy