AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Understanding Data Influence With Differential Approximation

Haoru Tan, Sitong Wu, Xiuzhe Wu, Wang Wang, Bo Zhao, Zeke Xie, Gui-Song Xia, Xiaojuan Qi

TPAMI 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Data attribution and recourse · 4 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 self-influence instead of validation influence (ablation)Diff-In · Data cleaning CIFAR-100 (20%)precision: 84.3 -> 91.4Improved
use self-influence instead of validation influence (ablation)Diff-In · Data cleaning ImageNet-1K (20%)precision: 47.2 -> 53.6Improved
replace TracIN with Diff-In (self-influence)TracIN · Data cleaning Tiny-ImageNet (20%)precision: 76.3 -> 88.2Improved
replace TracIN with Diff-In (self-influence)TracIN · Data cleaning ImageNet-1K (20%)precision: 46.4 -> 53.6Improved
replace SSSE with Diff-InSSSE · Data deletion GSM8KAccuracy@1: 85.7 -> 89.9Improved
replace TracIn with Diff-InTracIn · Coreset selection Tiny-ImageNet (20%)Accuracy@1: 24.2 -> 26.6Improved
replace TracIn with Diff-InTracIn · Coreset selection ImageNet-1K (30%)Accuracy@1: 64.4 -> 68.3Improved
replace Moderate with Diff-InModerate · Coreset selection CIFAR-100 (70%)Accuracy@1: 71.8 -> 71.5Hurt
use all uniformly selected time steps instead of only the final one (ablation)Diff-In-F · CC12M zero-shot classification (60%)Acc@1: 34.4 -> 36.2Improved
replace CLIP-score selection with Diff-InCLIP-score · CC12M zero-shot classification (60%)Acc@1: 34.1 -> 36.2Improved
increase number of time steps m from 3 to 5 (ablation)Diff-In · CC12M zero-shot classificationAcc@1: 24.3 -> 25.8Improved
increase number of time steps m from 5 to 50 (ablation)Diff-In · CC12M zero-shot classificationAcc@1: 25.8 -> 26.5No clear effect

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