Skip to contentDEEKOYA AI CoursesFLAT 20% OFFEnroll Now
Deekoya AI

Attribution Through the Decoder: Prediction-Space Explanations for Structured Outputs

Zaer Bin Zaqy, Brinnae Bent · 17 Nov 2026 · arXiv

Abstract

Networks that predict angles, gaze directions, and rotations do not output these quantities directly: they emit raw coordinates that a fixed decoder, such as normalization, maps to the final prediction. Attribution methods are usually applied to the raw coordinates, so they can credit inputs for changes the decoder discards, such as rescaling a vector when only its direction matters. The resulting explanation can highlight inputs that have no effect on the prediction. We instead attribute through the decoder differential, which for normalization decoders reduces to a scaled tangent projection. We characterize invariance to output rescaling, show that every fixed smooth scalar target on a sphere has points of zero tangential sensitivity, and show that an endpoint-projected variant of Integrated Gradients can still attribute to scale-only inputs, whereas direct angular Integrated Gradients cannot. On synthetic tasks with analytically known relevance, ambient attribution assigns up to 72% of its mass to an irrelevant region. On MPIIGaze validation data, tangent Grad-CAM produces 44% greater prediction displacement under 10% pixel deletion. On PASCAL3D+ with a 6D rotation decoder, the

cs.CVcs.AI

Comments

No comments yet. Share your own experience below.

Visitor posts are moderated before they appear and are the opinion of their authors, not DEEKOYA AI. Names are not verified. See our privacy policy.