TokSuite: Measuring the Impact of Tokenizer Choice on Language Model Behavior
This work addresses a foundational gap in NLP by systematically isolating tokenization's impact, which is crucial for researchers and practitioners in language modeling.
The paper tackles the problem of understanding how tokenizer choice affects language model performance by introducing TokSuite, a collection of 14 models with different tokenizers trained identically and a new benchmark for measuring real-world perturbations, revealing novel insights into the benefits and shortcomings of various tokenizers.
Tokenizers provide the fundamental basis through which text is represented and processed by language models (LMs). Despite the importance of tokenization, its role in LM performance and behavior is poorly understood due to the challenge of measuring the impact of tokenization in isolation. To address this need, we present TokSuite, a collection of models and a benchmark that supports research into tokenization's influence on LMs. Specifically, we train fourteen models that use different tokenizers but are otherwise identical using the same architecture, dataset, training budget, and initialization. Additionally, we curate and release a new benchmark that specifically measures model performance subject to real-world perturbations that are likely to influence tokenization. Together, TokSuite allows robust decoupling of the influence of a model's tokenizer, supporting a series of novel findings that elucidate the respective benefits and shortcomings of a wide range of popular tokenizers.