Related papers: Exploiting Vocabulary Frequency Imbalance in Langu…

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…

Large language models (LLMs) trained with canonical tokenization exhibit surprising robustness to non-canonical inputs such as character-level tokenization, yet the mechanisms underlying this robustness remain unclear. We study this…

Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution is to continually pre-train these models, saving significant…

State-of-the-art language models are autoregressive and operate on subword units known as tokens. Specifically, one must encode the conditioning string into a list of tokens before passing to the language models for next-token prediction.…

As the context length that large language models can handle continues to increase, these models demonstrate an enhanced ability to utilize distant information for tasks such as language modeling. This capability contrasts with human reading…

Increasing the size of large language models (LLMs) has been shown to lead to better performance. However, this comes at the cost of slower and more expensive inference. Early-exiting is a promising approach for improving the efficiency of…

Large language models (LLMs) have shown remarkable capabilities in many languages beyond English. Yet, LLMs require more inference steps when generating non-English text due to their reliance on English-centric tokenizers and vocabulary,…

Recurrent neural networks can learn to predict upcoming words remarkably well on average; in syntactically complex contexts, however, they often assign unexpectedly high probabilities to ungrammatical words. We investigate to what extent…

As Large Language Models (LLMs) scale, data curation has shifted from maximizing volume to optimizing the signal-to-noise ratio by performing quality filtering. However, for many languages, native high quality data is insufficient to train…

Traditionally, NLP performance improvement has been focused on improving models and increasing the number of model parameters. NLP vocabulary construction has remained focused on maximizing the number of words represented through subword…

One of the challenges with finetuning pretrained language models (PLMs) is that their tokenizer is optimized for the language(s) it was pretrained on, but brittle when it comes to previously unseen variations in the data. This can for…

Language models can largely benefit from efficient tokenization. However, they still mostly utilize the classical BPE algorithm, a simple and reliable method. This has been shown to cause such issues as under-trained tokens and sub-optimal…

Recent work on tokenizer-free multilingual pretrained models show promising results in improving cross-lingual transfer and reducing engineering overhead (Clark et al., 2022; Xue et al., 2022). However, these works mainly focus on reporting…

While many languages possess processes of joining two or more words to create compound words, previous studies have been typically limited only to languages with excessively productive compound formation (e.g., German, Dutch) and there is…

Neural networks are among the state-of-the-art techniques for language modeling. Existing neural language models typically map discrete words to distributed, dense vector representations. After information processing of the preceding…

State-of-the-art multilingual models depend on vocabularies that cover all of the languages the model will expect to see at inference time, but the standard methods for generating those vocabularies are not ideal for massively multilingual…

In this work we explore recent advances in Recurrent Neural Networks for large scale Language Modeling, a task central to language understanding. We extend current models to deal with two key challenges present in this task: corpora and…

Efficiency and safety of Large Language Models (LLMs), among other factors, rely on the quality of tokenization. A good tokenizer not only improves inference speed and language understanding but also provides extra defense against jailbreak…

Fine-tuning LLMs for classification typically maps inputs directly to labels. We ask whether attaching brief explanations to each label during fine-tuning yields better models. We evaluate conversational response quality along three axes:…

Why do larger language models generalize better? To investigate this question, we develop generalization bounds on the pretraining objective of large language models (LLMs) in the compute-optimal regime, as described by the Chinchilla…