AI Paper Atlas

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

Quantifying Learnability and Describability of Visual Concepts Emerging in Representation Learning

Iro Laina, Ruth Fong, Andrea Vedaldi

NeurIPS 2020 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded visual concept learning · 3 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 hard negatives from the most similar other class instead of random negativesHuman learnability evaluation of SeLa and MoCo clusters · Self-supervised classes, purity groupssemantic coherence: SeLa: 71.8 -> 55.3 (purity (0.3, 0.4]); MoCo: 89.8 -> 56.8 (purity (0.3, 0.4])Hurt
Select class captions using both within-class and other-class captionsClass-level caption generator · Google search, all classesBinary accuracy (%): 89.48 ± 0.07 -> 91.18 ± 0.07Improved
Drop the negative term from caption selection (ablation)Class-level caption generator · Google search, all classesTop-1 accuracy (%): 7.88 -> 7.64Hurt
Use ROUGE-L word-level distance instead of sentence-embedding distance (ablation)Class-level caption generator · Google search, all classesTop-1 accuracy (%): 7.88 -> 6.12Hurt
Use a random image as search query instead of a class-level captionClass description evaluation · Google search, all classesTop-1 accuracy (%): 7.88 -> 4.83Hurt
Use a representative image as search query instead of a class-level captionClass description evaluation · Google search, all classesTop-1 accuracy (%): 7.88 -> 4.05Hurt
Increase cluster puritySeLa and MoCo clusters · Self-supervised classes, random negativessemantic coherence: SeLa: 71.8 -> 98.8; MoCo: 89.8 -> 99.2Improved
Use automatic rather than human-written class descriptionsSeLa and MoCo clusters · Self-supervised classesdescribability: SeLa purity (0.3, 0.4]: 84.2 -> 78.7; MoCo purity (0.3, 0.4]: 78.3 -> 83.7Mixed

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