Summary of Interpreting Latent Student Knowledge Representations in Programming Assignments, by Nigel Fernandez et al.
Interpreting Latent Student Knowledge Representations in Programming Assignments
by Nigel Fernandez, Andrew Lan
First submitted to arxiv on: 13 May 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Computers and Society (cs.CY); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents InfoOIRT, an Information regularized Open-ended Item Response Theory model that uses generative large language models to predict open-ended student responses. The authors leverage recent advances in AI for education by focusing on interpreting the latent student knowledge representations learned by these models. By maximizing mutual information between a fixed subset of latent knowledge states and generated student code, InfoOIRT encourages disentangled representations of salient syntactic and semantic code features. The model is tested on a real-world programming education dataset, demonstrating its ability to accurately generate student code while leading to interpretable student knowledge representations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses special computers to understand how students learn programming skills. It tries to figure out what’s going through students’ minds when they write their own computer codes. The researchers created a new way of doing this called InfoOIRT, which helps the computer make sense of what it learned about students. They tested it on real student work and showed that it can accurately predict what students will write while also giving clues about how students think. |
