AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Large-Scale Few-Shot Learning via Multi-Modal Knowledge Discovery

Shuo Wang, Jun Yue, Jianzhuang Liu, Qi Tian, Meng Wang

ECCV 2020 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 39 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 semantic soft loss from textual knowledge (alpha 0.5) (ablation)LCE classifier · NOVEL-S1 K=1Top-5 accuracy: 49.5 -> 51.1Improved
add semantic soft loss from textual knowledge (alpha 0.5) (ablation)LCE classifier · NOVEL-S1 K=10Top-5 accuracy: 77.3 -> 78.8Improved
add semantic soft loss from textual knowledge (alpha 0.5) (ablation)LCE classifier · ALL-S1 K=20Top-5 accuracy: 79.3 -> 80.3Improved
add semantic soft loss, alpha sweep 0.0 to 0.9 (ablation)LCE classifier · NOVEL-S1 K=1Top-5 accuracy: 50.1 to 51.1 across alphaMixed
add imbalance control loss LIC (ablation)LCE classifier · NOVEL-S1 K=1Top-5 accuracy: 49.5 -> 50.1Improved
add imbalance control loss LIC (ablation)LCE classifier · NOVEL-S1 K=10Top-5 accuracy: 77.3 -> 78.1Improved
add imbalance control loss LIC (ablation)LCE classifier · ALL-S1 K=20Top-5 accuracy: 79.3 -> 79.9Improved
vary imbalance control strength beta in [0,1] (ablation)LCE + LIC · ImageNet-FSTop-5 accuracyNo clear effect
concatenate foreground and background features with original features (v) (ablation)Original-image CNN features · ImageNet-FS S1 (Fig 4)Top-5 accuracyImproved
add textual knowledge classifier to SGM featuresSGM · ImNet K=1Top-5 accuracy: 31.4 -> 33.5Improved
add textual knowledge classifier to KTCH featuresKTCH · ImNet K=1Top-5 accuracy: 36.0 -> 40.1Improved
add textual knowledge classifier to KTCH featuresKTCH · ImNet K=5Top-5 accuracy: 60.4 -> 63.3Improved

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