# Enhancing Science Classroom Discourse Analysis through Joint Multi-Task Learning for Reasoning-Component Classification

arXiv: https://arxiv.org/abs/2604.21137 · HTML: https://mlsanity.com/paper/2604.21137/

## Key facts

- No citing papers in the map yet (data through 14 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.

## Placed in

- LLM reasoning / [Mathematical reasoning](https://mlsanity.com/lr/mathematical-reasoning/)

## Builds on (most cited first, 12 of 12)

- [FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling](https://mlsanity.com/paper/2110.08263/) (2021)
- [FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning](https://mlsanity.com/paper/2205.07246/) (2022)
- [AugGPT: Leveraging ChatGPT for Text Data Augmentation](https://mlsanity.com/paper/2302.13007/) (2023)
- [Synthetic Data Generation with Large Language Models for Text Classification: Potential and Limitations](https://mlsanity.com/paper/2310.07849/) (2023)
- [AugESC: Dialogue Augmentation with Large Language Models for Emotional Support Conversation](https://mlsanity.com/paper/2202.13047/) (2022)
- [Debiased Self-Training for Semi-Supervised Learning](https://mlsanity.com/paper/2202.07136/) (2022)
- [Pretrained Language Models for Sequential Sentence Classification](https://mlsanity.com/paper/1909.04054/) (2019)
- [Generative and Discriminative Text Classification with Recurrent Neural Networks](https://mlsanity.com/paper/1703.01898/) (2017)
- [Hierarchical Neural Networks for Sequential Sentence Classification in Medical Scientific Abstracts](https://mlsanity.com/paper/1808.06161/) (2018)
- [Language Model Pre-training for Hierarchical Document Representations](https://mlsanity.com/paper/1901.09128/) (2019)
- [Enhancing Self-Training Methods](https://mlsanity.com/paper/2301.07294/) (2023)
- [Multi-label Sequential Sentence Classification via Large Language Model](https://mlsanity.com/paper/2411.15623/) (2024)

## 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.