Mechanistic Understanding and Mitigation of Language Confusion in English-Centric Large Language Models

MCML Authors

Ercong Nie

→ Group Hinrich Schütze
 Computational Linguistics

Hinrich Schütze

Prof. Dr.

Principal Investigator

Computational Linguistics

Abstract

Language confusion -- where large language models (LLMs) generate unintended languages against the user's need -- remains a critical challenge, especially for English-centric models. We present the first mechanistic interpretability (MI) study of language confusion, combining behavioral benchmarking with neuron-level analysis. Using the Language Confusion Benchmark (LCB), we show that confusion points (CPs) -- specific positions where language switches occur -- are central to this phenomenon. Through layer-wise analysis with TunedLens and targeted neuron attribution, we reveal that transition failures in the final layers drive confusion. We further demonstrate that editing a small set of critical neurons, identified via comparative analysis with multilingual-tuned models, substantially mitigates confusion without harming general competence or fluency. Our approach matches multilingual alignment in confusion reduction for most languages and yields cleaner, higher-quality outputs. These findings provide new insights into the internal dynamics of LLMs and highlight neuron-level interventions as a promising direction for robust, interpretable multilingual language modeling.

inproceedings

---

Findings @EMNLP 2025

Findings of the Conference on Empirical Methods in Natural Language Processing. Suzhou, China, Nov 04-09, 2025. To be published. Preprint available.

---

Authors

E. Nie • H. Schmid • H. Schütze

---

Links

---

Research Area

 B2 | Natural Language Processing