AI Paper Atlas

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

Evaluating Data Influence in Meta Learning

Chenyang Ren, Huanyi Xie, Shu Yang, Meng Ding, Dongrui Liu, Lijie Hu, Di Wang

ICLR 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Meta-augmentation and meta-objectives · 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
use task-IF (total gradient and Hessian) instead of direct IF (ablation)MAML task removal · omniglotaccuracy: 0.5422 -> 0.7804Improved
use task-IF instead of direct IF (ablation)MAML task removal · MNISTaccuracy: 0.3247 -> 0.6118Improved
use task-IF instead of direct IF (ablation)MAML task removal · MINI-Imagenetaccuracy: 0.2214 -> 0.2668Improved
use direct IF instead of task-IF (faster, less accurate) (ablation)MAML task removal · omniglotruntime (s): 65.69 -> 7.46Improved
use task-IF instead of retrainingMAML task removal · omniglotruntime (s): 316.74 -> 65.69Improved
use task-IF instead of retrainingMAML task removal · MINI-Imagenetaccuracy: 0.3071 -> 0.2668Hurt
use instance-IF (training) instead of retrainingMAML instance removal · omniglotruntime (s): 251.76 -> 4.49Improved
use instance-IF (validation) instead of retrainingMAML instance removal · MNISTaccuracy: 0.6265 -> 0.5476Hurt
remove tasks by influence score instead of randomlyMAML with 80% flipped-label tasks · MNIST harmful task removal, 25% checkedtest accuracy: 55 -> 62.5Improved
remove top-influence tasks instead of random tasksMAML · MNIST task leveltest accuracy: 0.62 -> 0.58Improved

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