ByT5: Towards a Token-Free Future with Pre-trained Byte-to-Byte Models (2022.tacl-1)
Copied to clipboard
Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel
| Challenge: | a number of pre-trained language models use sequences of tokens corresponding to word units . token-free models that operate directly on raw text have many advantages . |
| Approach: | They propose a standard Transformer architecture that can be used to process byte sequences . they also characterize trade-offs in terms of parameter count, training FLOPs, and inference speed . |
| Outcome: | The proposed model is more robust to noise and more robust on spelling and pronunciation tasks. |
Similar Papers
A Multi-dimensional Evaluation of Tokenizer-free Multilingual Pretrained Models (2023.findings-eacl)
Copied to clipboard
| Challenge: | Recent work on tokenizer-free models shows promising results in cross-lingual transfer . previous work focused on reporting accuracy on a limited set of tasks and data settings . |
| Approach: | They compare tokenizer-free and subword-based models using various dimensions . they find subword models are still the most practical choice in many settings . |
| Outcome: | The proposed model improves cross-lingual transfer and reduces engineering overhead. |
From Characters to Words: Hierarchical Pre-trained Language Model for Open-vocabulary Language Understanding (2023.acl-long)
Copied to clipboard
| Challenge: | Current models for natural language understanding require a preprocessing step to convert raw text into discrete tokens. |
| Approach: | They propose a hierarchical open-vocabulary language model that adopts a shallow Transformer architecture to learn word representations from their characters and a deep inter-word Transformer module that contextualizes each word representation by attending to the entire word sequence. |
| Outcome: | The proposed model outperforms baselines on various downstream tasks and is robust to textual corruption and domain shift. |
HashFormers: Towards Vocabulary-independent Pre-trained Transformers (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing pre-trained language models are vocabulary-dependent, mapping by default each token to its corresponding embedding. |
| Approach: | They propose a family of vocabulary-independent pre-trained transformers that support unlimited vocabulary . they propose to map each token to its corresponding embedding by default . |
| Outcome: | The proposed models are more memory efficient than existing models while achieving comparable performance on multiple text classification tasks. |
Byte Pair Encoding is Suboptimal for Language Model Pretraining (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Subword tokenization is a popular language model that can be used to segment text. |
| Approach: | They analyze differences between byte-pair encoding (BPE) and unigram LM tokenization methods to find subword units that align more closely with morphology. |
| Outcome: | The proposed method recovers subword units that align more closely with morphology and avoids problems stemming from BPE’s greedy construction procedure. |
Neural Machine Translation without Embeddings (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing models operate over subword tokens, but byte-based models employ a different approach . a one-hot representation of each byte does not hurt performance, but it improves BLEU scores . |
| Approach: | They propose to represent every computerized text as a sequence of bytes via UTF-8 . this eliminates the need for an embedding layer and improves performance . |
| Outcome: | The proposed model improves BLEU scores on byte-to-byte translation models compared to character-level models . the proposed model does not require an embedding layer and does not drop out of the decoder . |
What is the best recipe for character-level encoder-only modelling? (2023.acl-long)
Copied to clipboard
| Challenge: | aims to benchmark recent progress in language understanding models that output contextualised representations at the character level. |
| Approach: | They aim to find the best way to build and train character-level BERT-like models by comparing architectural innovations with pretraining objectives. |
| Outcome: | The proposed model outperforms a token-based model on a set of evaluation tasks with a fixed training procedure. |
Trainable, Multiword-aware Linguistic Tokenization Using Modern Neural Networks (2026.eacl-srw)
Copied to clipboard
| Challenge: | Tokenization is a fundamental task in natural language processing that forms the first step of many pipelines. |
| Approach: | They propose to use a standard tokenizer trained without MWE-awareness as a baseline and a character-level SRN+CRF model to train token-level models. |
| Outcome: | The proposed tokenizers are based on a character-level and token-level sequence labeling problem and are consistent with the proposed pipelines. |
Canine: Pre-training an Efficient Tokenization-Free Encoder for Language Representation (2022.tacl-1)
Copied to clipboard
| Challenge: | End-to-end neural models have replaced the traditional pipeline and require an explicit tokenization step. |
| Approach: | They propose a neural encoder that operates directly on character sequences without explicit tokenization or vocabulary and a pre-training strategy that optionally uses subwords as a soft inductive bias. |
| Outcome: | The proposed model outperforms a comparable mBert model on a multilingual benchmark by 5.7 F1 on the TyDi QA benchmark. |
Learn Your Tokens: Word-Pooled Tokenization for Language Modeling (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Language models typically tokenize text into subwords, using a deterministic, hand-engineered heuristic of combining characters into longer surface-level strings such as ‘ing’ or whole words. |
| Approach: | They propose a 'learn your tokens' scheme which pooles bytes/characters into word representations and decodes individual characters/bytes per word in parallel. |
| Outcome: | The proposed tokenizer outperforms subword models and byte/character models over the word boundary and outperformed on rare words by a factor of 30! |
Pre-trained Language Models Do Not Help Auto-regressive Text-to-Image Generation (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in image tokenizers have enabled text-to-image generation using auto-regressive methods, but these methods lack pre-trained language models for text-based models. |
| Approach: | They adapt a pre-trained language model for auto-regressive text-to-image generation and show that pre-train language models offer limited help. |
| Outcome: | The proposed model is compared with a pre-trained language model and shows that it is no more effective than random initialized models. |