Research map / Label-efficient and robust learning / Recognition and long-tail
I Am Big, You Are Little; I Am Right, You Are Wrong
David A. Kelly, Akchunya Chanchal, Nathan Blake
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 changed | Where | Measured | Effect |
|---|---|---|---|
| classification is incorrect rather than correct | MPS size across models · ImageNet-1k validation | MPS size increase (%): +2.6 | Hurt |
| classification is incorrect rather than correct | MPS size across models · Caltech-256 | MPS size increase (%): +3.3 | Hurt |
| use ConvNext-V2 Huge v1 instead of Inception V3 | model choice · ImageNet-1k | average MPS area: 0.239 -> 0.05 | Improved |
| use EVA Giant instead of Inception V4 | model choice · ImageNet-1k | average MPS area: 0.23 -> 0.054 | Improved |
| use ResNet152-D instead of ViT Huge V1 | model choice · ImageNet-1k | average MPS area: 0.156 -> 0.134 | Improved |
| compare architectures with Kruskal-Wallis test | MPS size · ImageNet-1k | H statistic: H(4) = 1176.134, p < 0.001 | Improved |
| compare models within the Inception family (Friedman test) | MPS size · ImageNet-1k | p-value: p=0.36 | No clear effect |
| compare models within ConvNext, EVA, ResNet, ViT (Friedman test) | MPS size · ImageNet-1k | p-value: p < 0.01 | Improved |
| use larger pretrained models (ViT Large vs ViT Huge V1) | MPS area · ImageNet-1k | average MPS area: 0.099 vs 0.156 | Mixed |
Closest papers by meaning
- When does dough become a bagel? Analyzing the remaining mistakes on ImageNet (NeurIPS 2022 · similarity 0.87)
- Few-Class Arena: A Benchmark for Efficient Selection of Vision Models and Dataset Difficulty Measurement (ICLR 2025 · similarity 0.86)
- An Empirical Evaluation of Current Convolutional Architectures' Ability to Manage Nuisance Location and Scale Variability (CVPR 2016 · similarity 0.86)
- SELECT: A Large-Scale Benchmark of Data Curation Strategies for Image Classification (NeurIPS 2024 · similarity 0.85)
- From Pixels to Perception: Interpretable Predictions via Instance-wise Grouped Feature Selection (ICML 2025 · similarity 0.85)
- Bayesian Exploration of Pre-trained Models for Low-shot Image Classification (CVPR 2024 · similarity 0.84)
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.
