Structural Retention Index (SRI): A Collapse Index Extension for Orthogonal Stability Assessment

This paper introduces the Structural Retention Index (SRI), a metric that quantifies how well AI systems preserve internal decision structure under perturbation. SRI measures stability dimensions orthogonal to the Collapse Index (CI), providing comprehensive dual-signal failure detection.

Key Results:

• Perfect complementarity: CI + SRI = 1.000 (inverse measures of the same stability phenomenon)
• Equal discriminative power: Both achieve AUC=0.874 for error detection
• Vastly outperforms confidence alone (AUC=0.171)
• Identifies hidden instability: 20 Type II cases (4.0%) show internal confidence shifts without visible label flips

Validation:
AG News 4-class text classification (500 base examples × 4 variants = 2,000 predictions). Reproducible dataset, generation pipeline, and validation metrics publicly available.

Methodology:
SRI computation remains proprietary to prevent adversarial optimization. Validation outputs are independently verifiable. This approach balances scientific transparency with IP protection, consistent with industry practices for evaluation frameworks.

Reproducibility:

• Dataset: agnews_ci_sri_demo.csv (2,000 rows)
• Generation script: generate_agnews_demo.py (MIT license)
• Validation script: validate_metrics.py (MIT license)
• GitHub: https://github.com/collapseindex/ci-sri

License: CC BY-NC-ND 4.0 (Attribution-NonCommercial-NoDerivatives)

Contact: ask@collapseindex.org | https://collapseindex.org