Enhancing Science Classroom Discourse Analysis through Joint Multi-Task Learning for Reasoning-Component Classification
On the map: LLM reasoning → Mathematical reasoning
Abstract (arXiv)
Analyzing the reasoning patterns of students in science classrooms is critical for understanding knowledge construction mechanism and improving instructional practice to maximize cognitive engagement, yet manual coding of classroom discourse at scale remains prohibitively labor-intensive. We present an automated discourse analysis system (ADAS) that jointly classifies teacher and student utterances along two complementary dimensions: Utterance Type and Reasoning Component derived from our prior CDAT framework. To address severe label imbalance among minority classes, we (1) stratify-resplit the annotated corpus, (2) apply LLM-based synthetic data augmentation targeting minority classes, and (3) train a dual-probe head RoBERTa-base classifier. A zero-shot GPT-5.4 baseline achieves macro-F1 of 0.467 on UT and 0.476 on RC, establishing meaningful upper bounds for prompt-only approaches motivating fine-tuning. Beyond classification, we conduct discourse pattern analyses including UTxRC co-occurrence profiling, Cognitive Complexity Index (CCI) computation per session, lag-sequential analysis, and IRF chain analysis, revealing that teacher Feedback-with-Question (Fq) moves are the most consistent antecedents of student inferential reasoning (SR-I). Our results demonstrate that LLM-based augmentation meaningfully improves UT minority-class recognition, and that the structural simplicity of the RC task makes it tractable even for lexical baselines.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Mathematical reasoning, one of 17 directions in LLM reasoning. Builds on 12 of its 33 references that are in the map. First posted on arXiv 22 Apr 2026.
Builds on
12 of 33 references.
References · 12
- 2021FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labelingcited by 537
- 2022FreeMatch: Self-adaptive Thresholding for Semi-supervised Learningcited by 232
- 2023AugGPT: Leveraging ChatGPT for Text Data Augmentationcited by 128
- 2023Synthetic Data Generation with Large Language Models for Text Classification: Potential and Limitationscited by 96
- 2022AugESC: Dialogue Augmentation with Large Language Models for Emotional Support Conversationcited by 79
- 2022Debiased Self-Training for Semi-Supervised Learningcited by 62
- 2019Pretrained Language Models for Sequential Sentence Classificationcited by 53
- 2017Generative and Discriminative Text Classification with Recurrent Neural Networkscited by 30
- 2018Hierarchical Neural Networks for Sequential Sentence Classification in Medical Scientific Abstractscited by 22
- 2019Language Model Pre-training for Hierarchical Document Representationscited by 7
- 2023Enhancing Self-Training Methodscited by 2
- 2024Multi-label Sequential Sentence Classification via Large Language Modelcited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Exploring Automated Recognition of Instructional Activity and Discourse from Multimodal Classroom Data
- 2026LLM Reasoning Predicts When Models Are Right: Evidence from Coding Classroom Discourse
- 2026Optimizing LLM Annotation of Classroom Discourse through Multi-Agent Orchestration
- 2026Can Multimodal LLMs See Science Instruction? Benchmarking Pedagogical Reasoning in K-12 Classroom Videos
- 2026EduBehaviors: Assertion-based Schemas for Auditable Coding of Educational Dialogues
- 2024Enhanced Classroom Dialogue Sequences Analysis with a Hybrid AI Agent: Merging Expert Rule-Base with Large Language Models
- 2025DIMSUM: Discourse in Mathematical Reasoning as a Supervision Module
- 2025EducationQ: Evaluating LLMs' Teaching Capabilities Through Multi-Agent Dialogue Framework