Differences between AI-generated and teacher assessments of laboratory reports may stem from difficulties retrieving and interpreting equations, graphs, tables, and units, not only from model reasoning capabilities.¶
Claim · All claims
Evidence · 1 study · designr2·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¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Discrepancies between model and instructor assessments are related not only to model capabilities but also to retrieval and interpretation of visual and mathematical elements in reports. → Marcos Abreu 2026
Evidence¶
Marcos Abreu 2026¶
Marcos Abreu, Cecilia Stari, Arturo C. Martí. (2026). AI-Assisted Assessment of Experimental Physics Laboratory Reports: Potential, Limitations, and Support for Teaching Practice. arXiv. https://arxiv.org/abs/2609.22417 doi:10.48550/arxiv.2609.22417
q2 · i? · design · r2
Analysis of an experience assessing Experimental Physics laboratory reports with ChatGPT models against instructor rubric assessments. The article reports that "differences between AI-generated and teacher assessments may be related not only to model capabilities but also to difficulties in retrieving and interpreting" report elements. No effect size is printed.
"The findings indicate that differences between AI-generated and teacher assessments may be related not only to model capabilities but also to difficulties in retrieving and interpreting equations, graphs, tables, units, and other visual elements."
Discussion¶
Related Claims¶
- AI pilots found models not yet capable of high-quality math output generation, including multiple choice problems, coherent word problems, and math visuals — related
- Students reported ChatGPT limitations in Quantum Theory learning: insufficient in-depth explanations, inability to interpret graphs or graphic equations, and occasional unexplainable mathematical errors — related
- LLM accuracy drops from 96% on text-only to 74% on multimodal physics problems, with significant differences emerging among models — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
ai-teacher-assessment-gaps-tied-to-evidence-retrieval |
title |
Differences between AI-generated and teacher assessments of laboratory reports may stem from difficulties retrieving and interpreting equations, graphs, tables, and units, not only from model reasoning capabilities. |
description |
Differences between AI-generated and teacher assessments of laboratory reports may stem from difficulties retrieving and interpreting equations, graphs, tables, and units, not only from model reasoning capabilities. |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — marcos-abreu-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.