Multi-word Tokenization for Sequence Compression
This addresses efficiency issues for industrial applications of LLMs, though it is incremental as it builds on existing tokenization methods.
The paper tackles the high computational cost of Large Language Models by introducing MWT, a Multi-Word Tokenizer that compresses sequences by representing frequent multi-word expressions as single tokens, resulting in increased performance with fixed sequence lengths and faster inference with negligible performance drops.
Large Language Models have proven highly successful at modelling a variety of tasks. However, this comes at a steep computational cost that hinders wider industrial uptake. In this paper, we present MWT: a Multi-Word Tokenizer that goes beyond word boundaries by representing frequent multi-word expressions as single tokens. MWTs produce a more compact and efficient tokenization that yields two benefits: (1) Increase in performance due to a greater coverage of input data given a fixed sequence length budget; (2) Faster and lighter inference due to the ability to reduce the sequence length with negligible drops in performance. Our results show that MWT is more robust across shorter sequence lengths, thus allowing for major speedups via early sequence truncation.