AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

DAVE: Distribution-Aware Attribution via ViT Gradient Decomposition

Adam Wróbel, Siddhartha Gairola, Jacek Tabor, Bernt Schiele, Bartosz Michał Zieliński, Dawid Damian Rymarczyk

ICML 2026 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 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
use DAVE instead of best competing methodViT-B/16 · ImageNet-1k localizationGridPG (%): +1.79Improved
use DAVE instead of best competing methodDeiT-B/16 · ImageNet-1k localizationGridPG (%): +8.05Improved
use DAVE instead of best competing methodDeiT-III-B/16 · ImageNet-1k localizationGridPG (%): +11.92Improved
use DAVE instead of best competing methodDINO ViT-B/16 · ImageNet-1k localizationGridPG (%): +1.35Improved
use DAVE instead of best competing method (C-LRP)ViT-B/16 · ImageNet-1k localizationEnergyPG (%): 80.82 -> 78.60Hurt
use DAVE instead of best competing methodDINO ViT-B/16 · ImageNet-1k localizationEnergyPG (%): +1.12Improved
use DAVE instead of inherent B-cos explanationB-cos-ViT · ImageNet-1k localization (B-cos models)GridPG (%): 79.67 -> 84.00Improved
use DAVE instead of inherent B-cos explanationB-cos-ViT · ImageNet-1k localization (B-cos models)EnergyPG (%): 69.41 -> 78.55Improved
use DAVE instead of inherent B-cos explanationB-cos-ViT-C · ImageNet-1k localization (B-cos models)GridPG (%): 87.66 -> 88.43Improved
use DAVE instead of inherent B-cos explanationB-cos-ViT-C · ImageNet-1k localization (B-cos models)EnergyPG (%): 75.61 -> 79.63Improved
increase neighborhood samples beyond 50DAVE · ImageNet-1k localizationGridPG/EnergyPGNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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