Python MCQ dataset for unlearning and relearning experiments¶
Product or Programme · All products and programmes
Evidence · 2 claims (2 for) · 1 study (1 design),q2· 0 of 1 report an effect size · 2 claims rest on one study
Description¶
A dataset of 2,074 web-collected and LLM-generated Python multiple-choice questions spanning 22 concept categories, created by Jiajia Song et al. for unlearning and relearning experiments with LLM-based teachable agents.
Components¶
Claims¶
- Machine unlearning induces progressively stronger novice-level performance in an LLM teachable agent, with accuracy and F1 declining as the unlearning ratio increases [+W]
- Dialogue evidence shows coach feedback induces genuine conceptual change, shifting the agent from misconception to conceptually correct explanation rather than superficial answer alignment [+W]
Related Products and Programmes¶
-
Key Sources¶
- Jiajia Song, Zhihan Guo, and Jionghao Lin. (2026). Simulating Novice Students Using Machine Unlearning and Relearning in Large Language Models. https://arxiv.org/abs/2603.26142
Page metadata
| Field | Value |
|---|---|
type |
product |
id |
python-mcq-dataset-for-unlearning-and-relearning-experiments |
title |
Python MCQ dataset for unlearning and relearning experiments |
description |
A dataset of 2,074 web-collected and LLM-generated Python multiple-choice questions spanning 22 concept categories, created by Jiajia Song et al. |
status |
review |
product_kind |
dataset |
generated |
by process:settle-candidates on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
Frontmatter is the page's machine-readable record — see the Schema & Guide.