AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

SCOUT: Cyclic Causal Discovery Under Soft Interventions with Unknown Targets

Alpar Turkoglu, Muralikrishnna Guruswamy Sethuraman, Faramarz Fekri

ICML 2026 · Probabilistic and scientific modeling › Causal discovery · approach: Interventional causal discovery · 0 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
add neural spline flow noise model (ablation)SCOUT · Synthetic target recovery, Shift, Gaussian noise (d=10)target recovery AUPRC: 0.144 -> 1.000Improved
add neural spline flow noise model (ablation)SCOUT · Synthetic target recovery, Shift, Exponential noise (d=10)target recovery AUPRC: 0.471 -> 1.000Improved
add neural spline flow noise model (ablation)SCOUT · Synthetic target recovery, Scale, Gaussian noise (d=10)target recovery AUPRC: 0.227 -> 1.000Improved
add neural spline flow noise model (ablation)SCOUT · Synthetic target recovery, Noise intervention, Gaussian noise (d=10)target recovery AUPRC: 0.393 -> 0.264Hurt
add neural spline flow noise model (ablation)SCOUT · Synthetic target recovery, Noise intervention, Gumbel noise (d=10)target recovery AUPRC: 0.463 -> 0.607Improved
increase graph size from d=10 to d=70BACKSHIFT · Shift target recovery, Gaussian noisetarget recovery AUPRC: 0.90 -> 0.03Hurt
increase graph size from d=10 to d=70SCOUT · Shift target recovery, Gaussian noisetarget recovery AUPRC: 1.00 -> 1.00No clear effect
use SCOUT instead of BACKSHIFT for noisy function interventionsSCOUT · Synthetic target recovery, Exponential noise (d=10)target recovery AUPRC: 0.313 -> 0.590Improved
use SCOUT instead of BACKSHIFT on scale interventions with Exponential noiseSCOUT · Synthetic target recovery, Exponential noise (d=10)target recovery AUPRC: 1.000 -> 0.975Hurt
use SCOUT instead of BACKSHIFT on SERGIOSCOUT · SERGIOtarget recovery AUPRC: 0.432 -> 0.978Improved

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