# Stable and Faithful Explanations for Knowledge Tracing

arXiv: https://arxiv.org/abs/2609.28502 · HTML: https://mlsanity.com/paper/2609.28502/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 66 references that are in the map.
- First posted on arXiv 15 Sep 2026.

## Builds on (most cited first, 30 of 30)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [Attention is not not Explanation](https://mlsanity.com/paper/1908.04626/) (2019)
- [Captum: A unified and generic model interpretability library for PyTorch](https://mlsanity.com/paper/2009.07896/) (2020)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [A Benchmark for Interpretability Methods in Deep Neural Networks](https://mlsanity.com/paper/1806.10758/) (2018)
- [Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods](https://mlsanity.com/paper/1911.02508/) (2019)
- [The many Shapley values for model explanation](https://mlsanity.com/paper/1908.08474/) (2019)
- [Deep Knowledge Tracing](https://mlsanity.com/paper/1506.05908/) (2015)
- [On the Robustness of Interpretability Methods](https://mlsanity.com/paper/1806.08049/) (2018)
- [Manipulating and Measuring Model Interpretability](https://mlsanity.com/paper/1802.07810/) (2018)
- [Explaining by Removing: A Unified Framework for Model Explanation](https://mlsanity.com/paper/2011.14878/) (2020)
- [Context-Aware Attentive Knowledge Tracing](https://mlsanity.com/paper/2007.12324/) (2020)
- [A Self-Attentive model for Knowledge Tracing](https://mlsanity.com/paper/1907.06837/) (2019)
- [Feature relevance quantification in explainable AI: A causal problem](https://mlsanity.com/paper/1910.13413/) (2019)
- [A Diagnostic Study of Explainability Techniques for Text Classification](https://mlsanity.com/paper/2009.13295/) (2020)
- [True to the Model or True to the Data?](https://mlsanity.com/paper/2006.16234/) (2020)
- [EdNet: A Large-Scale Hierarchical Dataset in Education](https://mlsanity.com/paper/1912.03072/) (2019)
- [S-LIME: Stabilized-LIME for Model Explanation](https://mlsanity.com/paper/2106.07875/) (2021)
- [pyKT: A Python Library to Benchmark Deep Learning based Knowledge Tracing Models](https://mlsanity.com/paper/2206.11460/) (2022)
- [simpleKT: A Simple But Tough-to-Beat Baseline for Knowledge Tracing](https://mlsanity.com/paper/2302.06881/) (2023)
- [Statistical stability indices for LIME: obtaining reliable explanations for Machine Learning models](https://mlsanity.com/paper/2001.11757/) (2020)
- [Evaluating the Explainers: Black-Box Explainable Machine Learning for Student Success Prediction in MOOCs](https://mlsanity.com/paper/2207.00551/) (2022)
- [Educational data mining and learning analytics: An updated survey](https://mlsanity.com/paper/2402.07956/) (2024)
- [Interpretable Knowledge Tracing: Simple and Efficient Student Modeling with Causal Relations](https://mlsanity.com/paper/2112.11209/) (2021)
- [Towards Interpretable Deep Learning Models for Knowledge Tracing](https://mlsanity.com/paper/2005.06139/) (2020)
- [Trusting the Explainers: Teacher Validation of Explainable Artificial Intelligence for Course Design](https://mlsanity.com/paper/2212.08955/) (2022)
- [Extending Deep Knowledge Tracing: Inferring Interpretable Knowledge and Predicting Post-System Performance](https://mlsanity.com/paper/1910.12597/) (2019)

## Abstract (arXiv)

Knowledge tracing (KT) models predict student performance opaquely, limiting pedagogical action. This study contributes a validation protocol testing predictive competitiveness (RQ1), explanation stability (RQ2) and retraining-based faithfulness (RQ3) together. Thirteen behavioral features across five pedagogical themes were engineered from ASSISTments 2009 and 2012, with history features computed from temporally preceding interactions and current response latency retained only for retrospective analysis. ASSISTments 2009 was rebuilt: the uncorrected skill-builder release duplicates each multi-skill interaction across one row per skill, and because those rows share one correctness label, they leak it into preceding-interaction features. Rebuilding lowered model AUC and reordered the explanation results. An Extreme Gradient Boosting (XGBoost) model explained with Tree SHapley Additive exPlanations (TreeSHAP) was compared against four deep baselines (DKT, SAKT, AKT and SimpleKT) under an information-matched protocol giving the deep models the same behavioral signals and restricting XGBoost to what is derivable from the identifier-and-correctness stream they consume. XGBoost reached an area under the curve (AUC) of 0.777 on 2012 and 0.786 on rebuilt 2009, with prediction-time AUCs of 0.771 and 0.775, respectively, after excluding current response latency; restricted to the baselines' information it performed as they did (0.697 against 0.700, and 0.717 against 0.720), locating the difference in information supplied, not model family. Rankings were consistent across folds, seeds and conditioning schemes (Spearman rho = 0.989-1.000), and removing top-ranked TreeSHAP features harmed AUC more than random removal, though split-gain and permutation rankings performed comparably. Student-level examples are illustrative interpretations, not validated recommendations.