Concept word effects on language model learning differ by part of speech
Every Part Has Its Role: Part-of-Speech Effects in LLM Post-Training
Computation and LanguageArtificial Intelligence
Summary
Large language models usually learn by predicting the next word exactly, which doesn't always recognize that similar words can mean almost the same thing. The authors studied how adding similar words for nouns, verbs, and adjectives during training affects the model’s accuracy and usefulness. They found that changing concepts in verbs improved the model’s task performance most, while overall accuracy improved on news text regardless of word type. Interestingly, the best results on tasks didn’t always mean the model’s internal thinking changed the most, suggesting that where and how you add related words matters a lot.
What this means in practice
- •For natural language processing engineers: Improve training data design by focusing concept variation on verbs to enhance model task performance.
- •For ai platform developers: Optimize model training strategies by applying concept augmentation differently depending on part of speech to improve language understanding.
Authors
Laya Iyer, Chen Shani
Abstract
Large language models are typically trained with next-token prediction (NTP), which rewards predicting specific tokens rather than explicitly accounting for the semantic similarities between different words or expressions. Prior work has addressed this limitation by augmenting training data with minimally varying sentences that replace a target word with conceptually related alternatives, e.g., "The tired coach ran home" $\rightarrow$ {"The tired trainer ran home", "The tired instructor ran home", ...}. Such concept augmentation improves performance, but it remains unclear whether its effectiveness depends on where in the sentence the concept is expressed. We reproduce and extend prior work $\textit{beyond nouns}$ to verbs and adjectives, asking whether the part of speech (POS) carrying the concept signal affects perplexity and downstream performance. We find different patterns across evaluation settings: concept augmentation improves perplexity most consistently on news, independent of the POS used, whereas verbs provide the most reliable downstream gains. Despite these behavioral differences, representational analyses using CKA and linear probing show that the conditions yielding the largest downstream gains are not necessarily those that most alter the model's internal representations. Together, these results suggest that the effectiveness of concept augmentation depends on where and how conceptual information is introduced during training.