# Disentangling Learning from Judgment: Representation Learning for Open Response Analytics

arXiv: https://arxiv.org/abs/2512.23941 · HTML: https://mlsanity.com/paper/2512.23941/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 32 references that are in the map.
- First posted on arXiv 30 Dec 2025.

## Builds on (most cited first, 6 of 6)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotations](https://mlsanity.com/paper/2110.05719/) (2021)
- [Survey on Automated Short Answer Grading with Deep Learning: from Word Embeddings to Transformers](https://mlsanity.com/paper/2204.03503/) (2022)
- [Can Large Language Models Match Tutoring System Adaptivity? A Benchmarking Study](https://mlsanity.com/paper/2504.05570/) (2025)
- [Automatic Short Answer Grading and Feedback Using Text Mining Methods](https://mlsanity.com/paper/1807.10543/) (2018)
- [Benchmarking Educational LLMs with Analytics: A Case Study on Gender Bias in Feedback](https://mlsanity.com/paper/2511.08225/) (2025)

## Built on by (most cited first, 5 of 5)

- [Representation Learning to Study Temporal Dynamics in Tutorial Scaffolding](https://mlsanity.com/paper/2603.24535/) (2026)
- [Understanding Teacher Revisions of Large Language Model-Generated Feedback](https://mlsanity.com/paper/2603.27806/) (2026)
- [Multimodal Analytics of Cybersecurity Crisis Preparation Exercises: What Predicts Success?](https://mlsanity.com/paper/2603.28553/) (2026)
- [Physiological and Semantic Patterns in Medical Teams Using an Intelligent Tutoring System](https://mlsanity.com/paper/2603.29950/) (2026)
- [Toward Measuring AI's Effects on Skill Formation: The Stock-Formation Gap](https://mlsanity.com/paper/2605.16283/) (2026)

## Abstract (arXiv)

Open-ended responses are central to learning, yet automated scoring often conflates what students wrote with how teachers grade. We present an analytics-first framework that separates content signals from rater tendencies, making judgments visible and auditable via analytics. Using de-identified ASSISTments mathematics responses, we model teacher histories as dynamic priors and represent text with sentence embeddings. We apply centroid normalization and response-problem embedding differences, and explicitly model teacher effects with priors to reduce problem- and teacher-related confounds. Temporally-validated linear models quantify the contributions of each signal, and model disagreements surface observations for qualitative inspection. Results show that teacher priors heavily influence grade predictions; the strongest results arise when priors are combined with content embeddings (AUC~0.815), while content-only models remain above chance but substantially weaker (AUC~0.626). Adjusting for rater effects sharpens the selection of features derived from content representations, retaining more informative embedding dimensions and revealing cases where semantic evidence supports understanding as opposed to surface-level differences in how students respond. The contribution presents a practical pipeline that transforms embeddings from mere features into learning analytics for reflection, enabling teachers and researchers to examine where grading practices align (or conflict) with evidence of student reasoning and learning.