# Alignment Drift in CEFR-prompted LLMs for Interactive Spanish Tutoring

arXiv: https://arxiv.org/abs/2505.08351 · HTML: https://mlsanity.com/paper/2505.08351/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 71 references that are in the map.
- First posted on arXiv 13 May 2025.

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

- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Let the LLMs Talk: Simulating Human-to-Human Conversational QA via Zero-Shot LLM-to-LLM Interactions](https://mlsanity.com/paper/2312.02913/) (2023)
- [Stateful Large Language Model Serving with Pensieve](https://mlsanity.com/paper/2312.05516/) (2023)
- [minicons: Enabling Flexible Behavioral and Representational Analyses of Transformer Language Models](https://mlsanity.com/paper/2203.13112/) (2022)
- [Text Readability Assessment for Second Language Learners](https://mlsanity.com/paper/1906.07580/) (2019)
- [EuroBERT: Scaling Multilingual Encoders for European Languages](https://mlsanity.com/paper/2503.05500/) (2025)
- [Flesch or Fumble? Evaluating Readability Standard Alignment of Instruction-Tuned Language Models](https://mlsanity.com/paper/2309.05454/) (2023)
- [From Tarzan to Tolkien: Controlling the Language Proficiency Level of LLMs for Content Generation](https://mlsanity.com/paper/2406.03030/) (2024)
- [LLM Roleplay: Simulating Human-Chatbot Interaction](https://mlsanity.com/paper/2407.03974/) (2024)
- [TextDescriptives: A Python package for calculating a large variety of metrics from text](https://mlsanity.com/paper/2301.02057/) (2023)
- [Beyond Flesch-Kincaid: Prompt-based Metrics Improve Difficulty Classification of Educational Texts](https://mlsanity.com/paper/2405.09482/) (2024)

## Built on by (most cited first, 7 of 7)

- [Toward Beginner-Friendly LLMs for Language Learning: Controlling Difficulty in Conversation](https://mlsanity.com/paper/2506.04072/) (2025)
- [German4All -- A Dataset and Model for Readability-Controlled Paraphrasing in German](https://mlsanity.com/paper/2508.17973/) (2025)
- [Steer Model beyond Assistant: Controlling System Prompt Strength via Contrastive Decoding](https://mlsanity.com/paper/2601.06403/) (2026)
- [Are the LLMs Capable of Maintaining at Least the Language Genus?](https://mlsanity.com/paper/2510.21561/) (2025)
- [Multilingual Embedding Probes Fail to Generalize Across Learner Corpora](https://mlsanity.com/paper/2604.07095/) (2026)
- [Incentives Of EdTech: A Systematic Review Of EduNLP Research](https://mlsanity.com/paper/2606.13691/) (2026)
- [Do LLM Tutors Actually Teach? A Process-Centered Synthesis of the Tutoring Loop](https://mlsanity.com/paper/2610.15692/) (2026)

## Abstract (arXiv)

This paper investigates the potentials of Large Language Models (LLMs) as adaptive tutors in the context of second-language learning. In particular, we evaluate whether system prompting can reliably constrain LLMs to generate only text appropriate to the student's competence level. We simulate full teacher-student dialogues in Spanish using instruction-tuned, open-source LLMs ranging in size from 7B to 12B parameters. Dialogues are generated by having an LLM alternate between tutor and student roles with separate chat histories. The output from the tutor model is then used to evaluate the effectiveness of CEFR-based prompting to control text difficulty across three proficiency levels (A1, B1, C1). Our findings suggest that while system prompting can be used to constrain model outputs, prompting alone is too brittle for sustained, long-term interactional contexts - a phenomenon we term alignment drift. Our results provide insights into the feasibility of LLMs for personalized, proficiency-aligned adaptive tutors and provide a scalable method for low-cost evaluation of model performance without human participants.