AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Generating Visual Explanations

Lisa Anne Hendricks, Zeynep Akata, Marcus Rohrbach, Jeff Donahue, Bernt Schiele, Trevor Darrell

ECCV 2016 · Vision-language and multimodal › Human and embodied modeling · approach: Abstract visual reasoning benchmarks · 558 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
add discriminative class loss (with label conditioning) (ablation)Explanation model · CUB-200 bird explanationsCIDEr: 44.7 -> 56.7Improved
add discriminative class loss (with label conditioning) (ablation)Explanation model · CUB-200 bird explanationsclass similarity: 40.86 -> 52.25Improved
add discriminative class loss (with label conditioning) (ablation)Explanation model · CUB-200 bird explanationsclass rank (lower better): 17.69 -> 13.12Improved
add label conditioning (with discriminative loss) (ablation)Explanation model · CUB-200 bird explanationsCIDEr: 51.9 -> 56.7Improved
add label conditioning (with discriminative loss) (ablation)Explanation model · CUB-200 bird explanationsclass rank (lower better): 19.80 -> 13.12Improved
add label conditioning (with discriminative loss) (ablation)Explanation model · CUB-200 bird explanationsbird expert rank (lower better): 3.22 -> 2.78Improved
add discriminative loss to label-only model (ablation)Explanation-Label · CUB-200 bird explanationsMETEOR: 28.1 -> 29.2Improved
add class label conditioning to description modelDescription baseline · CUB-200 bird explanationsCIDEr: 42.0 -> 44.7Improved
add discriminative loss to Description-style model without labelDescription baseline · CUB-200 bird explanationsCIDEr: 42.0 -> 51.9Improved
condition on class label (definition vs description)Description baseline · CUB-200 bird explanationsclass rank (lower better): 24.43 -> 15.82Improved
use explanation model instead of description for sentence-classifier accuracyDescription baseline · CUB-200 bird explanationssentence classifier accuracy (%): 22.32% -> 59.13%Improved

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