AI Paper Atlas

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

Why Can't I Open My Drawer? Mitigating Object-Driven Shortcuts in Zero-Shot Compositional Action Recognition

Geo Ahn, Inwoong Lee, Taeoh Kim, Minho Shim, Dongyoon Wee, Jinwoo Choi

ECCV 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Robust fine-tuning of zero-shot models · 1 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 CPR and TORC to C2C (ablation)C2C · Sth-com val (CLIP)unseen composition accuracy: 30.14 -> 35.16Improved
add TORC alone (ablation)C2C · Sth-com val (CLIP)verb@unseen-comp: 54.02 -> 57.49Improved
add CPR alone (ablation)C2C · Sth-com val (CLIP)unseen vs seen composition accuracy: unseen 30.14 -> 33.22; seen 47.31 -> 43.09Mixed
add L_cos temporal-reversal term (ablation)TORC · Sth-com val (CLIP)unseen composition accuracy: 30.14 -> 31.11Improved
add entropy term on shuffled features (ablation)TORC · Sth-com val (CLIP)unseen composition accuracy: 30.14 -> 30.84Improved
use both L_cos and L_ent (ablation)TORC · Sth-com val (CLIP)unseen composition accuracy: 30.14 -> 33.92Improved
optimize cross-entropy over the full verb-object space instead of batch-adaptive (ablation)CPR · Sth-com val (CLIP)unseen composition accuracy: 33.05 -> 17.34Hurt
use batch-adaptive label expansion instead of closed-world labels (ablation)CPR · Sth-com val (CLIP)unseen composition accuracy: 30.80 -> 33.05Improved
penalize all hard negatives instead of only frequent ones (ablation)CPR · Sth-com val (CLIP)composition H.M.: 37.67 -> 31.33Hurt
penalize frequent hard negatives with L_CPR (ablation)CPR · Sth-com val (CLIP)composition H.M.: 37.28 -> 37.67Improved
apply RCORE vs C2C on Sth-comC2C · Sth-com test (CLIP)unseen composition accuracy: 30.08 -> 33.90Improved
apply RCORE vs C2C on EK100-comC2C · EK100-com test (CLIP)unseen composition accuracy: 21.56 -> 28.41Improved

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