Puls Dane publiczne, aktualizowane automatycznie
arXiv – nowe publikacje publikacja

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

[Widzenie komputerowe] Autorzy: Zaer Bin Zaqy, Brinnae Bent

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 Gradient…

Kategoria arxiv
cs.CV
Kategoria
Widzenie komputerowe
Pdf
https://arxiv.org/pdf/2611.23161v1

Otwórz oryginał w źródle ↗

Więcej z tego źródła

Wszystkie