@inproceedings{gombert-etal-2026-rubrics,
title = "Rubrics as Semantic Subspaces: A Unified Approach to Rubric-based Constructed Response Scoring across Short Answers and Essays",
author = "Gombert, Sebastian and
Hahn, Sonja and
Andersen, Nico and
Camus, Leon and
Sun, Zhifan and
Nguyen, Ngoc Nhu Hao and
Zehner, Fabian and
Cong, Longwei and
Mehler, Alexander and
Drachsler, Hendrik",
editor = "Kochmar, Ekaterina and
Alhafni, Bashar and
Bann{\`o}, Stefano and
Bexte, Marie and
Burstein, Jill and
Horbach, Andrea and
Laarmann-Quante, Ronja and
Tack, Anais and
Yaneva, Victoria and
Yuan, Zheng",
booktitle = "Proceedings of the 21st Workshop on Innovative Use of {NLP} for Building Educational Applications ({BEA} 2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.bea-1.44/",
doi = "10.18653/v1/2026.bea-1.44",
pages = "624--634",
ISBN = "979-8-89176-409-5",
abstract = "Rubrics are the primary reference for manual scoring of constructed responses, and there is growing interest in their use in automated scoring methodologies. In this work, we propose Aspect-Grounded Rubric{--}Answer Alignment (AGRAA), a rubric-based end-to-end scoring framework that models rubric descriptors as latent aspect spaces. Concretely, rubric descriptors are represented as low-dimensional subspaces derived from contextualised transformer embeddings, and student responses are scored according to how strongly their representations align with these rubric-induced spaces relative to the residual space outside them. This formulation provides a geometrically grounded interpretation of rubric-based scoring while enabling end-to-end training with standard transformer encoders. We introduce three distinct architectural variants and evaluate them on multiple short-answer and essay scoring datasets. Across these tasks, AGRAA achieves predictive performance highly competitive with strong neural and feature-based baselines. In addition, the framework yields interpretable intermediate representations that expose which rubric-defined aspects contribute to scoring decisions, enabling decision-aligned explanations grounded in rubric descriptors."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gombert-etal-2026-rubrics">
<titleInfo>
<title>Rubrics as Semantic Subspaces: A Unified Approach to Rubric-based Constructed Response Scoring across Short Answers and Essays</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sebastian</namePart>
<namePart type="family">Gombert</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sonja</namePart>
<namePart type="family">Hahn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nico</namePart>
<namePart type="family">Andersen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leon</namePart>
<namePart type="family">Camus</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhifan</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ngoc</namePart>
<namePart type="given">Nhu</namePart>
<namePart type="given">Hao</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabian</namePart>
<namePart type="family">Zehner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Longwei</namePart>
<namePart type="family">Cong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Mehler</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hendrik</namePart>
<namePart type="family">Drachsler</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 21st Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bashar</namePart>
<namePart type="family">Alhafni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stefano</namePart>
<namePart type="family">Bannò</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie</namePart>
<namePart type="family">Bexte</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jill</namePart>
<namePart type="family">Burstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrea</namePart>
<namePart type="family">Horbach</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ronja</namePart>
<namePart type="family">Laarmann-Quante</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anais</namePart>
<namePart type="family">Tack</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Victoria</namePart>
<namePart type="family">Yaneva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zheng</namePart>
<namePart type="family">Yuan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-409-5</identifier>
</relatedItem>
<abstract>Rubrics are the primary reference for manual scoring of constructed responses, and there is growing interest in their use in automated scoring methodologies. In this work, we propose Aspect-Grounded Rubric–Answer Alignment (AGRAA), a rubric-based end-to-end scoring framework that models rubric descriptors as latent aspect spaces. Concretely, rubric descriptors are represented as low-dimensional subspaces derived from contextualised transformer embeddings, and student responses are scored according to how strongly their representations align with these rubric-induced spaces relative to the residual space outside them. This formulation provides a geometrically grounded interpretation of rubric-based scoring while enabling end-to-end training with standard transformer encoders. We introduce three distinct architectural variants and evaluate them on multiple short-answer and essay scoring datasets. Across these tasks, AGRAA achieves predictive performance highly competitive with strong neural and feature-based baselines. In addition, the framework yields interpretable intermediate representations that expose which rubric-defined aspects contribute to scoring decisions, enabling decision-aligned explanations grounded in rubric descriptors.</abstract>
<identifier type="citekey">gombert-etal-2026-rubrics</identifier>
<identifier type="doi">10.18653/v1/2026.bea-1.44</identifier>
<location>
<url>https://aclanthology.org/2026.bea-1.44/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>624</start>
<end>634</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Rubrics as Semantic Subspaces: A Unified Approach to Rubric-based Constructed Response Scoring across Short Answers and Essays
%A Gombert, Sebastian
%A Hahn, Sonja
%A Andersen, Nico
%A Camus, Leon
%A Sun, Zhifan
%A Nguyen, Ngoc Nhu Hao
%A Zehner, Fabian
%A Cong, Longwei
%A Mehler, Alexander
%A Drachsler, Hendrik
%Y Kochmar, Ekaterina
%Y Alhafni, Bashar
%Y Bannò, Stefano
%Y Bexte, Marie
%Y Burstein, Jill
%Y Horbach, Andrea
%Y Laarmann-Quante, Ronja
%Y Tack, Anais
%Y Yaneva, Victoria
%Y Yuan, Zheng
%S Proceedings of the 21st Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-409-5
%F gombert-etal-2026-rubrics
%X Rubrics are the primary reference for manual scoring of constructed responses, and there is growing interest in their use in automated scoring methodologies. In this work, we propose Aspect-Grounded Rubric–Answer Alignment (AGRAA), a rubric-based end-to-end scoring framework that models rubric descriptors as latent aspect spaces. Concretely, rubric descriptors are represented as low-dimensional subspaces derived from contextualised transformer embeddings, and student responses are scored according to how strongly their representations align with these rubric-induced spaces relative to the residual space outside them. This formulation provides a geometrically grounded interpretation of rubric-based scoring while enabling end-to-end training with standard transformer encoders. We introduce three distinct architectural variants and evaluate them on multiple short-answer and essay scoring datasets. Across these tasks, AGRAA achieves predictive performance highly competitive with strong neural and feature-based baselines. In addition, the framework yields interpretable intermediate representations that expose which rubric-defined aspects contribute to scoring decisions, enabling decision-aligned explanations grounded in rubric descriptors.
%R 10.18653/v1/2026.bea-1.44
%U https://aclanthology.org/2026.bea-1.44/
%U https://doi.org/10.18653/v1/2026.bea-1.44
%P 624-634
Markdown (Informal)
[Rubrics as Semantic Subspaces: A Unified Approach to Rubric-based Constructed Response Scoring across Short Answers and Essays](https://aclanthology.org/2026.bea-1.44/) (Gombert et al., BEA 2026)
ACL
- Sebastian Gombert, Sonja Hahn, Nico Andersen, Leon Camus, Zhifan Sun, Ngoc Nhu Hao Nguyen, Fabian Zehner, Longwei Cong, Alexander Mehler, and Hendrik Drachsler. 2026. Rubrics as Semantic Subspaces: A Unified Approach to Rubric-based Constructed Response Scoring across Short Answers and Essays. In Proceedings of the 21st Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2026), pages 624–634, San Diego, California, USA. Association for Computational Linguistics.