AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

A Study on PAVE Specification for Learnware

Hao-Yu Shi, Zhi-Hao Tan, Zi-Chen Zhao, Yang Yu, Zhi-Hua Zhou

ICLR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors

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
fit model predictions p(y_hat|x) instead of data distribution p(y|x) for model vectors (ablation)PAVE · CV with corrupted learnwares (average)accuracy: 0.745 -> 0.887Improved
fit model predictions instead of data distribution (ablation)PAVE · CIFAR10 (corrupted learnwares)accuracy: 0.894 -> 0.890No clear effect
fit model predictions instead of data distribution (ablation)PAVE · GTSRB (corrupted learnwares)accuracy: 0.656 -> 0.987Improved
fit model predictions instead of data distribution (ablation)PAVE · EuroSAT (corrupted learnwares)accuracy: 0.742 -> 0.991Improved
fit model predictions instead of data distribution (ablation)PAVE · SVHN (corrupted learnwares)accuracy: 0.643 -> 0.970Improved
identify learnware with PAVE instead of RKME, then fine-tunePAVE · NLP beyond original functionality (average)average score: 0.659 -> 0.709Improved
identify learnware with PAVE instead of random selectionPAVE · NLP beyond original functionality (average)average score: 0.668 -> 0.709Improved
fine-tune identified learnware instead of fine-tuning RoBERTa-LPAVE · NLP beyond original functionality (average)average score: 0.699 -> 0.709Improved
PAVE instead of RKME on QQP accuracyPAVE · QQP (beyond original functionality)Acc: 0.696 -> 0.631Hurt
PAVE instead of RKME on RTE accuracyPAVE · RTE (beyond original functionality)Acc: 0.635 -> 0.560Hurt
identify learnware with PAVE instead of random on corrupted CV repositoryPAVE · CV with corrupted learnwares (average)average accuracy: 0.580 -> 0.887Improved
identify learnware with PAVE instead of fine-tuning ViT-L-14PAVE · CV with corrupted learnwares (average)average accuracy: 0.733 -> 0.887Improved

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