AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

ZeroC: A Neuro-Symbolic Model for Zero-shot Concept Recognition and Acquisition at Inference Time

Tailin Wu, Megan Tjandrasuwita, Zhengxuan Wu, Xuelin Yang, Kevin Liu, Rok Sosic, Jure Leskovec

NeurIPS 2022 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Robust fine-tuning of zero-shot models · 4 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 R-EBM relation term to composition (ablation)ZeroC · HD-Letterclassification accuracy: 62.5 -> 84.5Improved
add R-EBM relation term to composition (ablation)ZeroC · HD-Conceptclassification accuracy: 32.5 -> 70.5Improved
add R-EBM relation term to composition (ablation)ZeroC · HD-Letter+distractordetection IoU: 45.3 -> 72.5Improved
add HC-EBM concept term to composition (ablation)ZeroC · HD-Letterclassification accuracy: 67.0 -> 84.5Improved
add HC-EBM concept term to composition (ablation)ZeroC · HD-Concept+distractordetection IoU: 78.4 -> 84.7Improved
add L(pos-std) loss term (ablation)ZeroC · HD-Letterclassification accuracy: 43.6 -> 84.5Improved
add L(pos-std) loss term (ablation)ZeroC · HD-Letter+distractordetection IoU: 76.1 -> 72.5Hurt
add L(neg) loss term (ablation)ZeroC · HD-Letterclassification accuracy: 64.5 -> 84.5Improved
add L(neg) loss term (ablation)ZeroC · HD-Letter+distractordetection IoU: 60.0 -> 72.5Improved
add L(em) loss term (ablation)ZeroC · HD-Letterclassification accuracy: 81.5 -> 84.5Mixed
use full objective instead of only L(Improved) (ablation)ZeroC · HD-Letterclassification accuracy: 27.5 -> 84.5Improved
add L(pos-std) loss term (ablation)ZeroC1 -> ZeroC2 · Domain 1 2D parsinggraph isomorphism accuracy: 55.2 -> 72.7Improved

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