Modeling Programming Skills with Source Code Embeddings for Context-aware Exercise Recommendation Conference: LAK26: 16th International Learning Analytics and Knowledge Conference; April 27-May 01, 2026; Bergen, Norway LAK26: 16th International Learning Analytics and Knowledge Conference (LAK 2026),
Carlos Eduardo P. Silva, João Pedro M. Sena, Julio C. S. Reis, André G. Santos, Lucas N. Ferreira · 2026
Read Modeling Programming Skills with Source Code Embeddings for Context-aware Exercise Recommendation Conference: LAK26: 16th International Learning Analytics and Knowledge Conference; April 27-May 01, 2026; Bergen, Norway LAK26: 16th International Learning Analytics and Knowledge Conference (LAK 2026), by Carlos Eduardo P. Silva, João Pedro M. Sena, Julio C. S. Reis, André G. Santos, Lucas N. Ferreira (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-10249.txt.
Description
Picking homework that fits a student's current skills requires knowing those skills across programming topics. A 2026 paper led by Carlos Eduardo P. Silva, with four co-authors, offers a context-aware recommender that models programming skills from code embeddings students submit during a course. Predicted skill profiles are matched to the skills unseen homework problems require, ranking exercises by how closely student profiles match problem skill vectors.
Tests on real data from a first programming course weigh embedding types for skill prediction, with Jina embeddings topping TF-IDF, CodeBERT-cpp, and GraphCodeBERT on most skills. A second test across seven course offerings checks recommendation fit with weekly content, finding the approach steadily more fitting than baselines using correctness or solution time. Predicted skills prove the stronger signal for problem picks.
Catalog record
| CREATOR | Carlos Eduardo P. Silva, João Pedro M. Sena, Julio C. S. Reis, André G. Santos, Lucas N. Ferreira |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-10249