Datasets:
scenario_id string | pressure float64 | buffer_capacity float64 | coupling_strength float64 | trajectory_drift float64 | perturbation_size float64 | perturbation_response float64 | stability_margin float64 | label_resilient_system int64 |
|---|---|---|---|---|---|---|---|---|
cpr_train_001 | 0.41 | 0.8 | 0.34 | -0.1 | 0.12 | 0.09 | 0.61 | 1 |
cpr_train_002 | 0.46 | 0.75 | 0.38 | -0.06 | 0.15 | 0.12 | 0.55 | 1 |
cpr_train_003 | 0.52 | 0.69 | 0.43 | -0.01 | 0.19 | 0.18 | 0.46 | 1 |
cpr_train_004 | 0.57 | 0.64 | 0.48 | 0.03 | 0.23 | 0.29 | 0.36 | 0 |
cpr_train_005 | 0.62 | 0.58 | 0.53 | 0.08 | 0.27 | 0.37 | 0.28 | 0 |
cpr_train_006 | 0.67 | 0.53 | 0.57 | 0.12 | 0.31 | 0.44 | 0.21 | 0 |
cpr_train_007 | 0.44 | 0.77 | 0.36 | -0.08 | 0.13 | 0.1 | 0.58 | 1 |
cpr_train_008 | 0.5 | 0.71 | 0.41 | -0.03 | 0.18 | 0.16 | 0.49 | 1 |
cpr_train_009 | 0.6 | 0.6 | 0.51 | 0.06 | 0.25 | 0.33 | 0.31 | 0 |
cpr_train_010 | 0.65 | 0.55 | 0.56 | 0.11 | 0.3 | 0.41 | 0.23 | 0 |
Clinical Perturbation Resilience Sepsis Detection
Overview
This dataset tests whether a model can detect whether a sepsis-like clinical system remains stable under perturbation.
Many systems appear stable at baseline but destabilize under relatively small shocks. The key question is whether the system can absorb disturbance and remain within the same stability regime, or whether perturbation pushes it toward collapse.
The goal of this benchmark is to determine whether the system is truly resilient or only conditionally stable.
Prediction target
label_resilient_system
0 = system destabilizes under perturbation
1 = system absorbs perturbation and remains stable
The task is to determine whether the system can maintain stability under disturbance.
Row structure
Each row represents a synthetic clinical scenario.
Columns:
scenario_id
pressure
buffer_capacity
coupling_strength
trajectory_drift
perturbation_size
perturbation_response
stability_margin
Training rows include the label.
Tester rows omit the label.
Evaluation
The scoring script reports:
accuracy
precision
recall
f1
specificity
negative predictive value (npv)
Primary metric
recall
Secondary metric
f1
Recall is prioritized because correctly identifying fragile systems under perturbation is critical for preventing collapse.
Why this benchmark matters
Clinical systems are often evaluated under nominal conditions, yet many failures occur when the system is exposed to additional stress.
A resilient system absorbs perturbation and returns to its recovery path. A fragile system shifts toward instability under the same disturbance.
This benchmark tests whether models can reason about shock absorption and regime persistence in dynamical systems.
Structural note
This dataset exposes system geometry while keeping the generator used to produce the scenarios private.
The goal is to evaluate whether models can detect structural resilience rather than memorizing surface stability patterns.
Clarus Stability Geometry Benchmarks
This dataset is part of a broader benchmark family exploring instability and recovery in complex systems.
Related probes include:
clinical-compensation-collapse-sepsis-v1
clinical-fork-point-sepsis-transition-v1
clinical-organ-failure-cascade-v1
clinical-recovery-window-sepsis-v1
clinical-intervention-alignment-sepsis-v1
clinical-recovery-stability-sepsis-v1
clinical-false-stability-sepsis-v1
clinical-instability-margin-sepsis-v1
clinical-intervention-competition-sepsis-v1
clinical-oscillatory-instability-sepsis-v1
clinical-counterfactual-intervention-sepsis-v1
clinical-intervention-timing-sepsis-v1
Together these benchmarks map the lifecycle of instability and recovery in clinical dynamical systems.
License
MIT
- Downloads last month
- 32