Individual and Combined Effects of English as a Second Language and Typos on LLM Performance

TL;DR

This paper investigates how ESL variations and typos jointly affect large language model performance, revealing that their combined impact often exceeds individual effects, especially on closed-ended tasks.

Contribution

It introduces a framework to systematically analyze the combined effects of ESL and typos on LLMs, highlighting the importance of realistic evaluation scenarios.

Findings

Combined ESL and typos cause larger performance drops than either factor alone.

Performance degradation is more consistent on closed-ended tasks.

Evaluations on clean English may overestimate real-world model performance.

Abstract

Large language models (LLMs) are used globally, and because much of their training data is in English, they typically perform best on English inputs. As a result, many non-native English speakers interact with them in English as a second language (ESL), and these inputs often contain typographical errors. Prior work has largely studied the effects of ESL variation and typographical errors separately, even though they often co-occur in real-world use. In this study, we use the Trans-EnV framework to transform standard English inputs into eight ESL variants and apply MulTypo to inject typos at three levels: low, moderate, and severe. We find that combining ESL variation and typos generally leads to larger performance drops than either factor alone, though the combined effect is not simply additive. This pattern is clearest on closed-ended tasks, where performance degradation can be characterized more consistently across ESL variants and typo levels, while results on open-ended tasks are more mixed. Overall, these findings suggest that evaluations on clean standard English may overestimate real-world model performance, and that evaluating ESL variation and typographical errors in isolation does not fully capture model behavior in realistic settings.