AI Paper Atlas

Research map / Video / Action recognition

Task-Specific Distance Correlation Matching for Few-Shot Action Recognition

Fei Long, Yao Zhang, Jiaming Lv, Jiangtao Xie, Peihua Li

AAAI 2026 · Video › Action recognition · approach: Disentangled action recognition · 0 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 Ladder Side Network to zero-shot CLIP (ablation)CLIP ViT-B/16 · SSv2-Full 1-shotaccuracy (%): 37.0 -> 67.1Improved
add IF-DaC on top of LSN (ablation)TS-FSAR · SSv2-Full 1-shotaccuracy (%): 67.1 -> 71.4Improved
add TSM on top of LSN+IF-DaC (ablation)TS-FSAR · SSv2-Full 1-shotaccuracy (%): 71.4 -> 73.8Improved
add GLAC on top of LSN+IF-DaC+TSM (ablation)TS-FSAR · HMDB51 5-shotaccuracy (%): 85.6 -> 88.9Improved
use query-specific task prototype (average) (ablation)TS-FSAR · SSv2-Full 1-shotaccuracy (%): 74.1 -> 75.1Improved
use averaging instead of cross-attention for query-specific prototype (ablation)TS-FSAR · SSv2-Full 1-shotaccuracy (%): 74.1 -> 75.1Improved
use alpha-DC instead of cosine similarity for inter-frame dependency (ablation)TS-FSAR · SSv2-Full 1-shotaccuracy (%): 71.5 -> 75.1Improved
use TS-DCM instead of BiMHM (ablation)TS-FSAR · SSv2-Full 1-shotaccuracy (%): 72.3 -> 75.1Improved
add IF-DaC to GAP metric (ablation)GAP · SSv2-Full 1-shotaccuracy (%): 68.6 -> 72.0Improved
use 12-layer LSN instead of 3-layer (ablation)TS-FSAR · SSv2-Full 1-shotaccuracy (%): 67.0 -> 75.1Improved
TS-FSAR versus TSAM on Kinetics-100TS-FSAR · Kinetics-100 5-shotaccuracy (%): 97.1 -> 96.6Hurt
TS-FSAR versus TSAM on SSv2-FullTS-FSAR · SSv2-Full 1-shotaccuracy (%): 65.8 -> 75.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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