# Can LLMs Simulate L2-English Dialogue? An Information-Theoretic Analysis of L1-Dependent Biases

arXiv: https://arxiv.org/abs/2502.14507 · HTML: https://mlsanity.com/paper/2502.14507/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 64 references that are in the map.
- First posted on arXiv 20 Feb 2025.

## Builds on (most cited first, 20 of 20)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [A Survey on In-context Learning](https://mlsanity.com/paper/2301.00234/) (2022)
- [Scaling Synthetic Data Creation with 1,000,000,000 Personas](https://mlsanity.com/paper/2406.20094/) (2024)
- [Large Language Models are few(1)-shot Table Reasoners](https://mlsanity.com/paper/2210.06710/) (2022)
- [Finding Support Examples for In-Context Learning](https://mlsanity.com/paper/2302.13539/) (2023)
- [Better to Ask in English: Cross-Lingual Evaluation of Large Language Models for Healthcare Queries](https://mlsanity.com/paper/2310.13132/) (2023)
- [Controllable Text Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2408.12599/) (2024)
- [LLM-as-a-tutor in EFL Writing Education: Focusing on Evaluation of Student-LLM Interaction](https://mlsanity.com/paper/2310.05191/) (2023)
- [Language Models as Models of Language](https://mlsanity.com/paper/2408.07144/) (2024)
- [ClarQ-LLM: A Benchmark for Models Clarifying and Requesting Information in Task-Oriented Dialog](https://mlsanity.com/paper/2409.06097/) (2024)
- [Embedding structure matters: Comparing methods to adapt multilingual vocabularies to new languages](https://mlsanity.com/paper/2309.04679/) (2023)
- [SLABERT Talk Pretty One Day: Modeling Second Language Acquisition with BERT](https://mlsanity.com/paper/2305.19589/) (2023)
- [LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users](https://mlsanity.com/paper/2406.17737/) (2024)
- [xDial-Eval: A Multilingual Open-Domain Dialogue Evaluation Benchmark](https://mlsanity.com/paper/2310.08958/) (2023)
- [Identifying the Correlation Between Language Distance and Cross-Lingual Transfer in a Multilingual Representation Space](https://mlsanity.com/paper/2305.02151/) (2023)
- [Second Language Acquisition of Neural Language Models](https://mlsanity.com/paper/2306.02920/) (2023)
- [Context-Aware LLM Translation System Using Conversation Summarization and Dialogue History](https://mlsanity.com/paper/2410.16775/) (2024)
- [A Three-Pronged Approach to Cross-Lingual Adaptation with Multilingual LLMs](https://mlsanity.com/paper/2406.17377/) (2024)
- [Interaction Matters: An Evaluation Framework for Interactive Dialogue Assessment on English Second Language Conversations](https://mlsanity.com/paper/2407.06479/) (2024)
- [An Interpretable and Crosslingual Method for Evaluating Second-Language Dialogues](https://mlsanity.com/paper/2408.16518/) (2024)

## Built on by (most cited first, 1 of 1)

- [Dango: A Strictly L1-Only Large Language Model for Studying Second Language Acquisition](https://mlsanity.com/paper/2606.19170/) (2026)

## Abstract (arXiv)

This study evaluates Large Language Models' (LLMs) ability to simulate non-native-like English use observed in human second language (L2) learners interfered with by their native first language (L1). In dialogue-based interviews, we prompt LLMs to mimic L2 English learners with specific L1s (e.g., Japanese, Thai, Urdu) across seven languages, comparing their outputs to real L2 learner data. Our analysis examines L1-driven linguistic biases, such as reference word usage and avoidance behaviors, using information-theoretic and distributional density measures. Results show that modern LLMs (e.g., Qwen2.5, LLAMA3.3, DeepseekV3, GPT-4o) replicate L1-dependent patterns observed in human L2 data, with distinct influences from various languages (e.g., Japanese, Korean, and Mandarin significantly affect tense agreement, and Urdu influences noun-verb collocations). Our results reveal the potential of LLMs for L2 dialogue generation and evaluation for future educational applications.