# Toward Beginner-Friendly LLMs for Language Learning: Controlling Difficulty in Conversation

arXiv: https://arxiv.org/abs/2506.04072 · HTML: https://mlsanity.com/paper/2506.04072/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 59 references that are in the map.
- First posted on arXiv 4 Jun 2025.

## Builds on (most cited first, 10 of 10)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference](https://mlsanity.com/paper/2412.13663/) (2024)
- [FUDGE: Controlled Text Generation With Future Discriminators](https://mlsanity.com/paper/2104.05218/) (2021)
- [Two Tales of Persona in LLMs: A Survey of Role-Playing and Personalization](https://mlsanity.com/paper/2406.01171/) (2024)
- [On the application of Large Language Models for language teaching and assessment technology](https://mlsanity.com/paper/2307.08393/) (2023)
- [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)
- [Kani: A Lightweight and Highly Hackable Framework for Building Language Model Applications](https://mlsanity.com/paper/2309.05542/) (2023)
- [Controlled Language Generation for Language Learning Items](https://mlsanity.com/paper/2211.15731/) (2022)
- [Alignment Drift in CEFR-prompted LLMs for Interactive Spanish Tutoring](https://mlsanity.com/paper/2505.08351/) (2025)

## Built on by (most cited first, 3 of 3)

- [EVALUESTEER: Measuring Reward Model Steerability Towards Values and Preferences](https://mlsanity.com/paper/2510.06370/) (2025)
- [Steer Model beyond Assistant: Controlling System Prompt Strength via Contrastive Decoding](https://mlsanity.com/paper/2601.06403/) (2026)
- [Multilingual Embedding Probes Fail to Generalize Across Learner Corpora](https://mlsanity.com/paper/2604.07095/) (2026)

## Abstract (arXiv)

Practicing conversations with large language models (LLMs) presents a promising alternative to traditional in-person language learning. However, most LLMs generate text at a near-native level of complexity, making them ill-suited for first and second-year beginner learners (CEFR: A1-A2). In this paper, we investigate whether controllable generation techniques can adapt LLM outputs to better support beginners. We evaluate these methods through both automatic metrics and a user study with university-level learners of Japanese. Our findings show that while prompting alone fails, controllable generation techniques can successfully improve output comprehensibility for beginner speakers (from 39.4% to 83.3%). We further introduce a new token-level evaluation metric, Token Miss Rate (TMR), that quantifies the proportion of incomprehensible tokens per utterance and correlates strongly with human judgments. To support future research in AI-assisted language learning, we release our code, models, annotation tools, and dataset.