@inproceedings{kaldaras-nalla-2026-learning,
title = "Learning Progression-Guided Scientific Model Assessment Using Foundation and Supervised Vision Models",
author = "Kaldaras, Leonora and
Nalla, Yasasvi",
editor = "Wilson, Joshua and
Ormerod, Christopher and
Beiting-Parrish, Magdalen",
booktitle = "Proceedings of the Artificial Intelligence in Measurement and Education Conference ({AIME}-Con): Coordinated Session Papers",
month = oct,
year = "2026",
address = "Wyndham Grand Pittsburgh Downtown, Pittsburgh, Pennsylvania, United States",
publisher = "National Council on Measurement in Education (NCME)",
url = "https://aclanthology.org/2026.aimecon-sessions.14/",
pages = "118--123",
ISBN = "979-8-9983004-2-4",
abstract = "Artificial intelligence can scale assessment of student-generated scientific models, but valid educational use requires algorithms to identify features that meaningfully represent the intended learning construct. This study compares foundation and supervised computer vision approaches for evaluating learning progression (LP)-aligned evidence in approximately 1,600 high-school students' electroscope models. Grounding DINO combined with the Segment Anything Model (SAM) was used for zero-shot detection without task-specific labeled training data, whereas a CustomCharge convolutional neural network (CNN) was trained on human-scored models. Both approaches were evaluated against expert scoring across 13 LP-aligned analytic categories. Grounding DINO+SAM achieved higher Cohen{'}s $\kappa$ in 12 of 13 categories, with strong performance across both charge- and force-related evidence. Categories involving less frequent or more complex cross-scenario representations remained comparatively challenging. Findings demonstrate complementary strengths of foundation and supervised approaches and illustrate how LPs can provide a theoretically grounded framework for developing and validating AI assessment of scientific models."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kaldaras-nalla-2026-learning">
<titleInfo>
<title>Learning Progression-Guided Scientific Model Assessment Using Foundation and Supervised Vision Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Leonora</namePart>
<namePart type="family">Kaldaras</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yasasvi</namePart>
<namePart type="family">Nalla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Artificial Intelligence in Measurement and Education Conference (AIME-Con): Coordinated Session Papers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Joshua</namePart>
<namePart type="family">Wilson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Ormerod</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Magdalen</namePart>
<namePart type="family">Beiting-Parrish</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>National Council on Measurement in Education (NCME)</publisher>
<place>
<placeTerm type="text">Wyndham Grand Pittsburgh Downtown, Pittsburgh, Pennsylvania, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-9983004-2-4</identifier>
</relatedItem>
<abstract>Artificial intelligence can scale assessment of student-generated scientific models, but valid educational use requires algorithms to identify features that meaningfully represent the intended learning construct. This study compares foundation and supervised computer vision approaches for evaluating learning progression (LP)-aligned evidence in approximately 1,600 high-school students’ electroscope models. Grounding DINO combined with the Segment Anything Model (SAM) was used for zero-shot detection without task-specific labeled training data, whereas a CustomCharge convolutional neural network (CNN) was trained on human-scored models. Both approaches were evaluated against expert scoring across 13 LP-aligned analytic categories. Grounding DINO+SAM achieved higher Cohen’s ąppa in 12 of 13 categories, with strong performance across both charge- and force-related evidence. Categories involving less frequent or more complex cross-scenario representations remained comparatively challenging. Findings demonstrate complementary strengths of foundation and supervised approaches and illustrate how LPs can provide a theoretically grounded framework for developing and validating AI assessment of scientific models.</abstract>
<identifier type="citekey">kaldaras-nalla-2026-learning</identifier>
<location>
<url>https://aclanthology.org/2026.aimecon-sessions.14/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>118</start>
<end>123</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Learning Progression-Guided Scientific Model Assessment Using Foundation and Supervised Vision Models
%A Kaldaras, Leonora
%A Nalla, Yasasvi
%Y Wilson, Joshua
%Y Ormerod, Christopher
%Y Beiting-Parrish, Magdalen
%S Proceedings of the Artificial Intelligence in Measurement and Education Conference (AIME-Con): Coordinated Session Papers
%D 2026
%8 October
%I National Council on Measurement in Education (NCME)
%C Wyndham Grand Pittsburgh Downtown, Pittsburgh, Pennsylvania, United States
%@ 979-8-9983004-2-4
%F kaldaras-nalla-2026-learning
%X Artificial intelligence can scale assessment of student-generated scientific models, but valid educational use requires algorithms to identify features that meaningfully represent the intended learning construct. This study compares foundation and supervised computer vision approaches for evaluating learning progression (LP)-aligned evidence in approximately 1,600 high-school students’ electroscope models. Grounding DINO combined with the Segment Anything Model (SAM) was used for zero-shot detection without task-specific labeled training data, whereas a CustomCharge convolutional neural network (CNN) was trained on human-scored models. Both approaches were evaluated against expert scoring across 13 LP-aligned analytic categories. Grounding DINO+SAM achieved higher Cohen’s ąppa in 12 of 13 categories, with strong performance across both charge- and force-related evidence. Categories involving less frequent or more complex cross-scenario representations remained comparatively challenging. Findings demonstrate complementary strengths of foundation and supervised approaches and illustrate how LPs can provide a theoretically grounded framework for developing and validating AI assessment of scientific models.
%U https://aclanthology.org/2026.aimecon-sessions.14/
%P 118-123
Markdown (Informal)
[Learning Progression-Guided Scientific Model Assessment Using Foundation and Supervised Vision Models](https://aclanthology.org/2026.aimecon-sessions.14/) (Kaldaras & Nalla, AIME-Con 2026)
ACL