AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Hierarchical Causal Abduction: A Foundation Framework for Explainable Model Predictive Control

Ramesh Arvind Naagarajan, Zühal Wagner, Stefan Streif

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Counterfactual and statistical explanations · 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 knowledge graph evidence (ablation)HCA · GreenhouseAnswer Correctness: 0.302 -> 0.478Improved
add PCMCI causal discovery (ablation)HCA · GreenhouseAnswer Correctness: 0.324 -> 0.478Improved
add KKT optimization evidence (ablation)HCA · GreenhouseAnswer Correctness: 0.325 -> 0.478Improved
add knowledge graph evidence (ablation)HCA · TEPAnswer Correctness: 0.393 -> 0.406No clear effect
add PCMCI causal discovery (ablation)HCA · TEPAnswer Correctness: 0.415 -> 0.406Hurt
add KG, PCMCI and KKT together instead of any single source (ablation)HCA · Building HVACAnswer Correctness: 0.201 -> 0.394Improved
use HCA instead of LIMEexplanation method · GreenhouseAnswer Correctness: 0.311 -> 0.478Improved
domain-specific KKT-threshold calibration (2-3 days)HCA · three domainsAnswer Correctness: 0.478 -> 0.88Improved
thresholds calibrated on Building data instead of Greenhouse thresholdsHCA · Building HVACAnswer Correctness: 0-1% AC lossNo clear effect
use Greenhouse-calibrated thresholds on TEPHCA · TEPAnswer Correctness: 19-21% AC lossHurt
use few-shot GPT-4o synthesis instead of template-based (ablation)HCA · 67 scenariosP@1: 0.710 -> 0.896Improved
remove 10-30% of knowledge graph edges (ablation)HCA · robustnessAC change: -12.5%Hurt

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