@inproceedings{arana-etal-2026-pathbuilder,
title = "{P}ath{B}uilder: A Quality-Controlled {LLM} System for Personalized Learning Pathways",
author = "Arana, Jasper Meynard and
Ma{\~n}acop, John Andrew and
Manacop, John Allen and
Garcia, Roy Andrew and
Piniera, Keith Rick and
Carandang, Kristine Ann M. and
Casin, Ethan Robert and
Alis, Christian and
Monterola, Christopher",
editor = "Durrett, Greg and
Jian, Ping",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 3: System Demonstrations)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-demo.50/",
pages = "504--514",
ISBN = "979-8-89176-392-0",
abstract = "Large language models (LLMs) enable scalable content generation for personalized learning, but reliability and pedagogical alignment remain open challenges. We present PathBuilder, a web-based system that integrates expert-validated assessment, retrieval-augmented generation (RAG), and an LLM-as-a-Judge validation loop within a closed instructional pipeline. The system uses a 17,758-item curriculum-aligned question bank, including 1,018 expert-approved LLM-generated items, to construct diagnostic and post-tests for fine-grained learner profiling. In a real-world deployment with 179 registered users (75 matched learners), PathBuilder achieved a mean absolute gain of 37.9 percentage points, Hake{'}s normalized gain of 0.760, and a large effect size (Cohen{'}s d = 0.98). A controlled study of the judge mechanism showed consistent high-quality instructional outputs with a 100{\%} threshold pass rate. These results demonstrate that structured curriculum alignment combined with retrieval grounding and automated validation can support reliable LLM-based personalization in deployed learning systems. A live demonstration of PathBuilder is available at https://demo.pathbuilderedu.com."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="arana-etal-2026-pathbuilder">
<titleInfo>
<title>PathBuilder: A Quality-Controlled LLM System for Personalized Learning Pathways</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jasper</namePart>
<namePart type="given">Meynard</namePart>
<namePart type="family">Arana</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">John</namePart>
<namePart type="given">Andrew</namePart>
<namePart type="family">Mañacop</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">John</namePart>
<namePart type="given">Allen</namePart>
<namePart type="family">Manacop</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roy</namePart>
<namePart type="given">Andrew</namePart>
<namePart type="family">Garcia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Keith</namePart>
<namePart type="given">Rick</namePart>
<namePart type="family">Piniera</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kristine</namePart>
<namePart type="given">Ann</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Carandang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ethan</namePart>
<namePart type="given">Robert</namePart>
<namePart type="family">Casin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christian</namePart>
<namePart type="family">Alis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Monterola</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 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Greg</namePart>
<namePart type="family">Durrett</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ping</namePart>
<namePart type="family">Jian</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, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-392-0</identifier>
</relatedItem>
<abstract>Large language models (LLMs) enable scalable content generation for personalized learning, but reliability and pedagogical alignment remain open challenges. We present PathBuilder, a web-based system that integrates expert-validated assessment, retrieval-augmented generation (RAG), and an LLM-as-a-Judge validation loop within a closed instructional pipeline. The system uses a 17,758-item curriculum-aligned question bank, including 1,018 expert-approved LLM-generated items, to construct diagnostic and post-tests for fine-grained learner profiling. In a real-world deployment with 179 registered users (75 matched learners), PathBuilder achieved a mean absolute gain of 37.9 percentage points, Hake’s normalized gain of 0.760, and a large effect size (Cohen’s d = 0.98). A controlled study of the judge mechanism showed consistent high-quality instructional outputs with a 100% threshold pass rate. These results demonstrate that structured curriculum alignment combined with retrieval grounding and automated validation can support reliable LLM-based personalization in deployed learning systems. A live demonstration of PathBuilder is available at https://demo.pathbuilderedu.com.</abstract>
<identifier type="citekey">arana-etal-2026-pathbuilder</identifier>
<location>
<url>https://aclanthology.org/2026.acl-demo.50/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>504</start>
<end>514</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PathBuilder: A Quality-Controlled LLM System for Personalized Learning Pathways
%A Arana, Jasper Meynard
%A Mañacop, John Andrew
%A Manacop, John Allen
%A Garcia, Roy Andrew
%A Piniera, Keith Rick
%A Carandang, Kristine Ann M.
%A Casin, Ethan Robert
%A Alis, Christian
%A Monterola, Christopher
%Y Durrett, Greg
%Y Jian, Ping
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-392-0
%F arana-etal-2026-pathbuilder
%X Large language models (LLMs) enable scalable content generation for personalized learning, but reliability and pedagogical alignment remain open challenges. We present PathBuilder, a web-based system that integrates expert-validated assessment, retrieval-augmented generation (RAG), and an LLM-as-a-Judge validation loop within a closed instructional pipeline. The system uses a 17,758-item curriculum-aligned question bank, including 1,018 expert-approved LLM-generated items, to construct diagnostic and post-tests for fine-grained learner profiling. In a real-world deployment with 179 registered users (75 matched learners), PathBuilder achieved a mean absolute gain of 37.9 percentage points, Hake’s normalized gain of 0.760, and a large effect size (Cohen’s d = 0.98). A controlled study of the judge mechanism showed consistent high-quality instructional outputs with a 100% threshold pass rate. These results demonstrate that structured curriculum alignment combined with retrieval grounding and automated validation can support reliable LLM-based personalization in deployed learning systems. A live demonstration of PathBuilder is available at https://demo.pathbuilderedu.com.
%U https://aclanthology.org/2026.acl-demo.50/
%P 504-514
Markdown (Informal)
[PathBuilder: A Quality-Controlled LLM System for Personalized Learning Pathways](https://aclanthology.org/2026.acl-demo.50/) (Arana et al., ACL 2026)
ACL
- Jasper Meynard Arana, John Andrew Mañacop, John Allen Manacop, Roy Andrew Garcia, Keith Rick Piniera, Kristine Ann M. Carandang, Ethan Robert Casin, Christian Alis, and Christopher Monterola. 2026. PathBuilder: A Quality-Controlled LLM System for Personalized Learning Pathways. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations), pages 504–514, San Diego, California, United States. Association for Computational Linguistics.