The visualizer's flagging mechanisms flagged all tested logs from cases already known to involve plagiarism while still requiring student-level review¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? In validation testing, all logs from cases already known to involve plagiarism were flagged by the application, and detailed student-level views enabled identification of the relevant cases. → Heidi Taveter and Marina Lepp 2026

Evidence¶

Heidi Taveter and Marina Lepp 2026¶

Heidi Taveter and Marina Lepp. (2026). From Idea to Classroom in Days: Using “Vibe Coding” to Create a Programming Process Visualizer from IDE Activity Logs. https://arxiv.org/abs/2607.24757 doi:10.48550/arxiv.2607.24757

q2 · i? · design · r2

Developer-led validation with diverse Thonny logs from previous courses, including logs from known plagiarism cases. The article reports that "All of these cases were flagged by the application", and that examining flagged entities through student-level views enabled identification of the relevant cases. No effect size or count of plagiarism cases is printed.

"The application was also tested with logs from cases already known to involve plagiarism. All of these cases were flagged by the application (Fig. 6)."

Discussion¶

Related Claims¶

• MOSS detected far fewer suspicious cases in Spring 2026 than in prior offerings, as LLM-assisted submissions lack similarity signals — related
• The programming-process replay was the most prompt-intensive view to develop, requiring 31 prompts to work reliably and handle edge cases — related
• Trustworthiness could not be reduced to a fully automated objective; human judgment remains necessary for resolving complex or context-dependent cases — a broader claim this one bears on