Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chen, Eason, Wang, Isabel, Yuan, Nina, Judicke, Sophia, Beigh, Kayla, Tang, Xinyi
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.27440
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917365820686336
author Chen, Eason
Wang, Isabel
Yuan, Nina
Judicke, Sophia
Beigh, Kayla
Tang, Xinyi
author_facet Chen, Eason
Wang, Isabel
Yuan, Nina
Judicke, Sophia
Beigh, Kayla
Tang, Xinyi
contents Behavioral analysis of tutoring dialogues is essential for understanding student learning, yet manual coding remains a bottleneck. We present a methodology where LLM coding agents autonomously improve the prompts used by LLM classifiers to label educational dialogues. In each iteration, a coding agent runs the classifier against human-labeled validation data, analyzes disagreements, and proposes theory-grounded prompt modifications for researcher review. Applying this approach to 659 AI tutoring sessions across four experiments with three agents and three classifiers, 4-fold cross-validation on held-out data confirmed genuine improvement: the best agent achieved test $κ=0.78$ (SD$=0.08$), matching human inter-rater reliability ($κ=0.78$), at a cost of approximately \$5--8 per agent. While development-set performance reached $κ=0.91$--$0.93$, the cross-validated results represent our primary generalization claim. The iterative process also surfaced an undocumented labeling pattern: human coders consistently treated expressions of confusion as engagement rather than disengagement. Continued iteration beyond the optimum led to regression, underscoring the need for held-out validation. We release all prompts, iteration logs, and data.
format Preprint
id arxiv_https___arxiv_org_abs_2603_27440
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Tool to Teammate: LLM Coding Agents as Collaborative Partners for Behavioral Labeling in Educational Dialogue Analysis
Chen, Eason
Wang, Isabel
Yuan, Nina
Judicke, Sophia
Beigh, Kayla
Tang, Xinyi
Human-Computer Interaction
Behavioral analysis of tutoring dialogues is essential for understanding student learning, yet manual coding remains a bottleneck. We present a methodology where LLM coding agents autonomously improve the prompts used by LLM classifiers to label educational dialogues. In each iteration, a coding agent runs the classifier against human-labeled validation data, analyzes disagreements, and proposes theory-grounded prompt modifications for researcher review. Applying this approach to 659 AI tutoring sessions across four experiments with three agents and three classifiers, 4-fold cross-validation on held-out data confirmed genuine improvement: the best agent achieved test $κ=0.78$ (SD$=0.08$), matching human inter-rater reliability ($κ=0.78$), at a cost of approximately \$5--8 per agent. While development-set performance reached $κ=0.91$--$0.93$, the cross-validated results represent our primary generalization claim. The iterative process also surfaced an undocumented labeling pattern: human coders consistently treated expressions of confusion as engagement rather than disengagement. Continued iteration beyond the optimum led to regression, underscoring the need for held-out validation. We release all prompts, iteration logs, and data.
title From Tool to Teammate: LLM Coding Agents as Collaborative Partners for Behavioral Labeling in Educational Dialogue Analysis
topic Human-Computer Interaction
url https://arxiv.org/abs/2603.27440