AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

PITE: Multi-Prototype Alignment for Individual Treatment Effect Estimation

Fuyuan Cao, Jiaxuan Zhang, Xiaoli Li

AAAI 2026 · Probabilistic and scientific modeling › Treatment effects · approach: Meta-learners and partial identification · 2 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
replace distribution-level alignment (CFR-Wass) with multi-prototype alignmentCFR-style ITE estimator · IHDP out-of-samplesqrt PEHE: 0.83(0.08) -> 0.60(0.04)Improved
replace instance-level alignment (CITE) with prototype alignmentCITE · IHDP out-of-samplesqrt PEHE: 0.67(0.14) -> 0.60(0.04)Improved
replace instance-level alignment (FCCL) with prototype alignmentFCCL · IHDP within-samplesqrt PEHE: 0.53(0.04) -> 0.51(0.02)Improved
use PITE instead of FCCLFCCL · IHDP within-sampleATE error: 0.09(0.01) -> 0.09(0.01)No clear effect
use PITE under higher covariate dispersionCFR-MMD · Synthetic gamma 0.4 -> 1.2sqrt PEHE: 2.54 -> 5.37 vs PITE 2.22 -> 4.99Improved
use prototype alignmentSITE · Synthetic gamma=1.2sqrt PEHE: 6.19(0.15) -> 4.99(0.10)Improved
prototype alignment improves representation uniformityPITE vs CFRNet · IHDPuniformity: 0.4734 -> 0.4566Improved

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