The Interlingua Hypothesis: LLMs Translate via a Latent Task-agnostic Feature Space

2026-08-31

AI Summary

研究探讨了大型语言模型通过潜在特征空间进行翻译的机制，提出并验证了中介语假设，揭示了模型在单语和翻译任务中的共通性。

Abstract

Large language models (LLMs) have recently demonstrated improved machine translation performance over strong supervised baselines. This raises questions as to what mechanisms underlie how LLMs perform machine translation between languages. Motivated by recent interpretability findings--namely, that LLMs use massively multilingual latent feature representations to perform language modeling--we propose the interlingua hypothesis. The hypothesis holds that language models translate by reading a source sentence into a latent feature space, and generate a target sentence by reading from the latent feature space. We show three lines of evidence in support of this hypothesis: (1) variance in BLEU across language pairs is largely predictable from language-specific competences with no language pair-specific interaction terms; (2) many model components are causally influential in both monolingual tasks and translation tasks; and (3) fine-tuning on monolingual data recovers a large proportion of translation improvements relative to fine-tuning on aligned documents. Together, these provide convergent evidence in support of the interlingua hypothesis, and suggest new ways of understanding and improving how LLMs can be leveraged to perform translation tasks.

Problem

• Large language models
• machine translation
• latent feature space
• interlingua hypothesis

Innovation

• interlingua hypothesis
• latent feature space
• machine translation