AI Paper Atlas

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

Universal Few-shot Learning of Dense Prediction Tasks with Visual Token Matching

Donggyun Kim, Jinwoo Kim, Seongwoong Cho, Chong Luo, Seunghoon Hong

ICLR 2023 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Fine-grained and cross-domain few-shot baselines · 3 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 VTM rather than the strongest fully supervised baselineTaskonomy SS, 10-shot · Taskonomy SSmIoU: 0.4449 -> 0.4097Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy SN, 10-shot · Taskonomy SNmErr: 6.4414 -> 11.4391Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy ED, 10-shot · Taskonomy EDRMSE: 0.0534 -> 0.0741Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy ZD, 10-shot · Taskonomy ZDRMSE: 0.0268 -> 0.0316Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy TE, 10-shot · Taskonomy TERMSE: 0.0188 -> 0.0791Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy OE, 10-shot · Taskonomy OERMSE: 0.0689 -> 0.0912Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy K2, 10-shot · Taskonomy K2RMSE: 0.0358 -> 0.0639Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy K3, 10-shot · Taskonomy K3RMSE: 0.0357 -> 0.0519Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy RS, 10-shot · Taskonomy RSRMSE: 0.0860 -> 0.1089Hurt
use VTM rather than the strongest fully supervised baselineTaskonomy PC, 10-shot · Taskonomy PCRMSE: 0.0347 -> 0.0420Hurt
add task-specific adaptation (ablation)VTM on Taskonomy SS · Taskonomy SSmIoU: 0.0002 -> 0.4097Improved
add non-parametric matching instead of a parametric linear mapping (ablation)VTM on Taskonomy SS · Taskonomy SSmIoU: 0.2681 -> 0.4097Improved

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