mlsanityData to 7 Oct 2026

Programming Knowledge Tracing: A Comprehensive Dataset and A New Model

arXiv 2112.08273 ↗First posted 11 Dec 2021cs.PL cs.AIcited by 9

Not on the map. Why?

Abstract (arXiv)

In this paper, we study knowledge tracing in the domain of programming education and make two important contributions. First, we harvest and publish so far the most comprehensive dataset, namely BePKT, which covers various online behaviors in an OJ system, including programming text problems, knowledge annotations, user-submitted code and system-logged events. Second, we propose a new model PDKT to exploit the enriched context for accurate student behavior prediction. More specifically, we construct a bipartite graph for programming problem embedding, and design an improved pre-training model PLCodeBERT for code embedding, as well as a double-sequence RNN model with exponential decay attention for effective feature fusion. Experimental results on the new dataset BePKT show that our proposed model establishes state-of-the-art performance in programming knowledge tracing. In addition, we verify that our code embedding strategy based on PLCodeBERT is complementary to existing knowledge tracing models to further enhance their accuracy. As a side product, PLCodeBERT also results in better performance in other programming-related tasks such as code clone detection.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 31 of its 53 references that are in the map. First posted on arXiv 11 Dec 2021.

Builds on

31 of 53 references.

References · 31

  1. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  2. 2017Graph Attention Networkscited by 8,854
  3. 2016node2vec: Scalable Feature Learning for Networkscited by 1,913
  4. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  5. 2020Don't Stop Pretraining: Adapt Language Models to Domains and Taskscited by 1,548
  6. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  7. 2020CodeBERT: A Pre-Trained Model for Programming and Natural Languagescited by 1,286
  8. 2014Convolutional Neural Networks for Sentence Classificationcited by 1,124
  9. 2018Efficient Neural Architecture Search via Parameter Sharingcited by 831
  10. 2021CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generationcited by 732
  11. 2020GraphCodeBERT: Pre-training Code Representations with Data Flowcited by 600
  12. 2021Unified Pre-training for Program Understanding and Generationcited by 361
19 more
  1. 2017Large-Scale Evolution of Image Classifierscited by 348
  2. 2015Deep Knowledge Tracingcited by 328
  3. 2020Revisiting Pre-Trained Models for Chinese Natural Language Processingcited by 262
  4. 2018code2vec: Learning Distributed Representations of Codecited by 249
  5. 2020IntelliCode Compose: Code Generation Using Transformercited by 225
  6. 2020Context-Aware Attentive Knowledge Tracingcited by 172
  7. 2016Dynamic Key-Value Memory Networks for Knowledge Tracingcited by 170
  8. 2019A Self-Attentive model for Knowledge Tracingcited by 169
  9. 2016Recurrent Neural Network for Text Classification with Multi-Task Learningcited by 144
  10. 2014Learning to Executecited by 105
  11. 2020Retrieval-Augmented Generation for Code Summarization via Hybrid GNNcited by 89
  12. 2021Language-Agnostic Representation Learning of Source Code from Structure and Contextcited by 63
  13. 2018Addressing Two Problems in Deep Knowledge Tracing via Prediction-Consistent Regularizationcited by 53
  14. 2020GIKT: A Graph-based Interaction Model for Knowledge Tracingcited by 51
  15. 2018Knowledge Tracing Machines: Factorization Machines for Knowledge Tracingcited by 42
  16. 2020Improving Knowledge Tracing via Pre-training Question Embeddingscited by 36
  17. 2018Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Informationcited by 28
  18. 2016A deep language model for software codecited by 14
  19. 2020HGKT: Introducing Hierarchical Exercise Graph for Knowledge Tracingcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 0; 2024: 3; 2025: 3; 2026: 2.

9 papers.

  1. 2024Test Case-Informed Knowledge Tracing for Open-ended Coding Taskscited by 7
  2. 2025CoderAgent: Simulating Student Behavior for Personalized Programming Learning with Large Language Modelscited by 4
  3. 2022GPT-based Open-Ended Knowledge Tracingcited by 3
  4. 2024Interpreting Latent Student Knowledge Representations in Programming Assignmentscited by 2
  5. 2026MERIT: Memory-Enhanced Retrieval for Interpretable Knowledge Tracingcited by 1
  6. 2024Personalized Programming Guidance based on Deep Programming Learning Style Capturingcited by 0
  7. 2025CIKT: A Collaborative and Iterative Knowledge Tracing Framework with Large Language Modelscited by 0
  8. 2025Denoising Programming Knowledge Tracing with a Code Graph-based Tuning Adaptorcited by 0
  9. 2026Using LLMs for Knowledge Component-level Correctness Labeling in Open-ended Coding Problemscited by 0