AI Paper Atlas

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

FALCON: Fast Visual Concept Learning by Integrating Images, Linguistic descriptions, and Conceptual Relations

Lingjie Mei, Jiayuan Mao, Ziqi Wang, Chuang Gan, Joshua B. Tenenbaum

ICLR 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded visual concept learning · 6 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 FALCON-G instead of NSCL+LSTMCUB fast concept learning · CUBQA accuracy: 79.53 -> 81.33Improved
Use box embeddings instead of hyperplane embeddings (ablation)FALCON-G · CUBQA accuracy: 74.11 -> 81.33Improved
Use box embeddings instead of hypercone embeddings (ablation)FALCON-G · CUBQA accuracy: 68.44 -> 81.33Improved
Add parsed concept graphs and GNN processing for supplemental sentences (ablation)NSCL+LSTM on CUB · CUBQA accuracy: 79.53 -> 78.50Improved
Replace averaged visual examples with an example-graph GNN (ablation)NSCL+GNN on CUB · CUBQA accuracy: 78.50 -> 81.33Improved
Train with fewer base concepts (ablation)FALCON-G · CUBQA accuracy: 81.33% -> 76.32%Hurt
Add supplemental sentences containing related concepts (ablation)FALCON-G · CUBQA accuracy: 76.37% -> 81.33%Improved
Use FALCON-G instead of NSCL+GNNCLEVR with supplemental descriptions · CLEVRQA accuracy: 84.11 -> 88.40Improved
Provide supplemental sentences rather than examples alone (ablation)FALCON-G · CLEVRQA accuracy: 68.47 -> 88.40Improved
Provide supplemental sentences rather than examples alone (ablation)CNN+LSTM · CLEVRQA accuracy: 41.92 -> 42.69No clear effect
Use FALCON-G instead of NSCL+GNNCLEVR biased-data learning · CLEVR biased dataaccuracy: 84.83% -> 88.29%Improved
Use FALCON instead of MACGQA with supplemental sentences · GQAaccuracy: 54.99% -> 55.96%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