AIBearisharXiv – CS AI · Apr 76/10
🧠
Individual and Combined Effects of English as a Second Language and Typos on LLM Performance
Research reveals that Large Language Models (LLMs) experience greater performance degradation when facing English as a Second Language (ESL) inputs combined with typographical errors, compared to either factor alone. The study tested eight ESL variants with three levels of typos, finding that evaluations on clean English may overestimate real-world model performance.