AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Matched Data, Better Models: Target Aligned Data Filtering with Sparse Features

Arnav Mohanty Das, Gantavya Bhatt, Sahil Verma, Yiping Wang, Viswa Virinchi Muppirala, Jeff Bilmes

ICLR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning

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 sparse SAE features instead of dense CLIP features (ablation)SDM · DataComp-medium ImageNet-1Kaccuracy: 24.1 -> 35.2Improved
use sparse SAE features instead of dense CLIP features (ablation)SDM · DataComp-medium 38-task averageaccuracy: 29.1 -> 36.4Improved
add concave log term (submodularity) (ablation)SDM · DataComp-medium 38-task averageaccuracy: 35.1 -> 36.4Improved
add concave log term (submodularity) (ablation)SDM · DataComp-medium ImageNet-1Kaccuracy: 34.6 -> 35.2Improved
add monotonicity loss L_mono to SAE training (ablation)SDM · DataComp-medium 38-task averageaccuracy: 35.00% -> 36.40%Improved
add monotonicity loss L_mono to SAE training (ablation)SDM · DataComp-medium ImageNet-1Kaccuracy: 34.80% -> 35.20%Improved
intersect 5 stochastic greedy runs instead of a single run (ablation)SDM · DataComp-medium ImageNet-1Kaccuracy: 33.5 -> 35.2Improved
use a broader 24-dataset target instead of ImageNet-1K (ablation)SDM · DataComp-medium 38-task averageaccuracy: 36.4 -> 35.8Hurt
use SDM instead of NCL ∩ NormSimNCL ∩ NormSim (IN1K) · DataComp-medium ImageNet-1Kaccuracy: 32.8% -> 35.2%Improved
use SDM instead of NCL ∩ NormSimNCL ∩ NormSim (Target) · DataComp-medium 38-task averageaccuracy: 35.7% -> 36.4%Improved
combine SDM with NCL + DFN + HYPENCL + DFN + HYPE · DataComp-medium 38-task averageaccuracy: 38.8% -> 39.2%Improved
combine SDM with NCL + DFN + HYPENCL + DFN + HYPE · DataComp-medium ImageNet-1Kaccuracy: 38.2% -> 39.2%Improved

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