AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Characterizing Underrepresentation in Generalizing Causal Survival Estimates

Bolun Liu, Sean McGrath, Yiren Hou, Elizabeth Stuart, Harsh Parikh

NeurIPS 2026 · Probabilistic and scientific modeling › Treatment effects · approach: Meta-learners and partial identification

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Identifies and characterizes target-population subgroups that are underrepresented in a randomized trial cohort when generalizing causal survival estimates.

Problem

Generalizing randomized trial findings to broader target populations in time-to-event settings can yield suboptimal decisions when the trial sample is misaligned with the target population.

Approach

Identifies and characterizes the target subgroups underrepresented in the trial cohort for time-to-event generalization, where underrepresentation arises in two ways (the abstract is truncated, so the second mechanism is not stated).

Measured effects reported in the paper

What was changedWhereMeasuredEffect
restrict generalization to reliably generalizable subgroups (refined estimand)TRACE on JTPA-to-CPS · JTPA 24-month employment effect95% CI width: [-10.5, +15.9] -> [-1.9, +8.6], 60.6% narrowerImproved
apply optimal trim B and censoring intervention hTRACE · JTPA-to-CPS generalizationvariance objective: 81.58 -> 12.68 (-84.5%)Improved
add R-gating to the censoring intervention (ablation)TRACE · JTPA-to-CPS generalizationh-active person-intervals: 1,796 -> 98Improved
compare TRACE against l-thresholding, censoring thresholding and ROOT on l aloneTRACE · Synthetic experimentsmechanism attributionImproved

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