# FMMO: Detecting the Divergence Between Local Attribution and Global Drift

arXiv: https://arxiv.org/abs/2609.06173 · HTML: https://mlsanity.com/paper/2609.06173/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 60 references that are in the map.
- First posted on arXiv 5 Sep 2026.

## Builds on (most cited first, 18 of 18)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods](https://mlsanity.com/paper/1911.02508/) (2019)
- [Consistent Individualized Feature Attribution for Tree Ensembles](https://mlsanity.com/paper/1802.03888/) (2018)
- [Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift](https://mlsanity.com/paper/1810.11953/) (2018)
- [A Random Forest Guided Tour](https://mlsanity.com/paper/1511.05741/) (2015)
- [Online Adaptation to Label Distribution Shift](https://mlsanity.com/paper/2107.04520/) (2021)
- [Amazon SageMaker Model Monitor: A System for Real-Time Insights into Deployed Machine Learning Models](https://mlsanity.com/paper/2111.13657/) (2021)
- [ConceptExplorer: Visual Analysis of Concept Driftsin Multi-source Time-series Data](https://mlsanity.com/paper/2007.15272/) (2020)
- [Tackling Virtual and Real Concept Drifts: An Adaptive Gaussian Mixture Model](https://mlsanity.com/paper/2102.05983/) (2021)
- [Confidence-based Estimators for Predictive Performance in Model Monitoring](https://mlsanity.com/paper/2407.08649/) (2024)
- [Evaluating Model Explanations without Ground Truth](https://mlsanity.com/paper/2505.10399/) (2025)
- [Explanation Shift: How Did the Distribution Shift Impact the Model?](https://mlsanity.com/paper/2303.08081/) (2023)
- [Unified Shapley Framework to Explain Prediction Drift](https://mlsanity.com/paper/2102.07862/) (2021)
- [Autoregressive based Drift Detection Method](https://mlsanity.com/paper/2203.04769/) (2022)
- [How to Sustainably Monitor ML-Enabled Systems? Accuracy and Energy Efficiency Tradeoffs in Concept Drift Detection](https://mlsanity.com/paper/2404.19452/) (2024)
- [Temporal Matrix Factorization for Tracking Concept Drift in Individual User Preferences](https://mlsanity.com/paper/1510.05263/) (2015)
- [A Scalable Approach to Covariate and Concept Drift Management via Adaptive Data Segmentation](https://mlsanity.com/paper/2411.15616/) (2024)

## Abstract (arXiv)

Post-deployment drift poses a critical risk to algorithmic accountability, particularly when ground truth labels are delayed and performance degradation becomes a "silent failure". While Explainable AI (XAI) is often relied upon to audit these shifts, we demonstrate that popular local attribution methods (e.g., TreeSHAP) can exhibit misleading stability even as model reliability collapses. In this paper, we propose a Framework for Model Monitoring and Observability (FMMO) designed to expose the divergence between local explanation stability and global distribution shifts. Using benchmark, synthetic, and real-world datasets, we show that local XAI methods fail to flag drift-induced disparate impact, specifically where False Positive Rates spike for protected groups while feature attributions remain unchanged. By integrating global surrogate models with model utilization measurements, FMMO mitigates this fairness blind spot, ensuring that stakeholders can detect discriminatory deterioration that standard local XAI tools overlook.