Artificial Intelligence

    

Tokenization is not the Problem

Authors: Danil Kutny

This paper introduces a modification to standard GPT-like models by incorporating character-level encoding. The model uses an LSTM to process individual characters within tokens, which are then embedded into the original token embedding space. This allows the model to maintain token-level processing while adding character-level information to each token. Trained on the BookCorpus dataset, the model was evaluated on tasks requiring character-level manipulation, such as counting letters and reversing words. Surprisingly, the modified model performed similarly to the baseline GPT model, with no significant improvements, suggesting that GPT-like models may inherently learn character-level representations from tokenized inputs.

Comments: 6 Pages.

Download: PDF

Submission history

[v1] 2024-12-22 14:24:26

Unique-IP document downloads: 225 times

Vixra.org is a pre-print repository rather than a journal. Articles hosted may not yet have been verified by peer-review and should be treated as preliminary. In particular, anything that appears to include financial or legal advice or proposed medical treatments should be treated with due caution. Vixra.org will not be responsible for any consequences of actions that result from any form of use of any documents on this website.

Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.

comments powered by Disqus