AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

f-INE: A Hypothesis Testing Framework for Estimating Influence under Training Randomness

Subhodip Panda, Dhruv Tarsadiya, Shashwat Sourav, Prathosh AP, Sai Praneeth Karimireddy

ICLR 2026 · Learning theory and algorithms › Sample complexity · approach: Optimal mean estimation and subsampling

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 hypothesis-testing f-INE influence instead of mean-based LESSf-INE · LIMA poisoning, Joe Biden, top 20% rankingrecall of poisoned instructions (%): 44% -> more than 60%Improved
compare full distributions (f-INE) instead of means (LESS)f-INE · LIMA poisoning variability across 3 runsaverage coefficient of variation: 64% lower than LESS at p=1.0Improved
f-INE vs TraceIn for mislabel detection on MNISTf-INE · MNIST mislabeled detectionrecall: +0.05% vs TraceInNo clear effect
f-INE vs TRAK and Influence Functions on MNIST mislabel detectionf-INE · MNIST mislabeled detectionrecall: +13.85% over TRAK, +3.83% over IFsImproved
decrease gradient projection dimension (from 8192 down to 1024) (ablation)f-INE · LIMA poisoningutility (recall of poisoned instructions)Hurt
use more checkpoints for gradient collection (60, 80 vs 100) (ablation)f-INE · LIMA poisoningutilityImproved

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