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.

Similar Papers

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.
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!
ByT5: Towards a Token-Free Future with Pre-trained Byte-to-Byte Models (2022.tacl-1)

Copied to clipboard

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.
On the Importance of Word Boundaries in Character-level Neural Machine Translation (D19-56)

Copied to clipboard

Challenge: Neural Machine Translation models typically use a fixed-size lexical vocabulary . subword segmentation methods rely on statistical heuristics that lack any linguistic notion .
Approach: They propose a hierarchical decoding architecture for character-level NMT using subwords . they propose fewer parameters and a more efficient approach to perform translation at the level of words .
Outcome: The proposed model can reach higher translation accuracy than the subword-level model with fewer parameters while maintaining longer-distance contextual and grammatical dependencies.
Unsupervised Morphological Tree Tokenizer (2025.findings-acl)

Copied to clipboard

Challenge: Conventional statistical tokenizers often disrupt constituent boundaries within words, thereby corrupting semantic information.
Approach: They propose a method that uses morphological structure guidance to induce character-level structures of words by training a deep model.
Outcome: Empirical results show that the proposed method retains complete morphemes and outperforms existing methods on morphological segmentation and language modeling tasks.
Towards Reasonably-Sized Character-Level Transformer NMT by Finetuning Subword Systems (2020.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to train character-level models require very deep architectures that are difficult and slow to train.
Approach: They propose to fine tune a Transformer token-based model to get a model without token segmentation.
Outcome: The proposed model improves translation quality and robustness to noise while requiring less token segmentation.
CharBERT: Character-aware Pre-trained Language Model (2020.coling-main)

Copied to clipboard

Challenge: Pre-trained language models (PLMs) construct word representations at subword level with Byte-Pair Encoding (BPE) or its variations . but these methods split a word into subword units and make it incomplete and fragile .
Approach: They propose a character-aware pre-trained language model to tackle OOV problems . they construct contextual word embedding for each token from sequential character representations .
Outcome: The proposed model improves on the existing models on multiple NLP benchmarks.
How Do Language Models Acquire Character-Level Information? (2026.eacl-long)

Copied to clipboard

Challenge: Language models (LMs) implicitly encode character-level information, despite not being explicitly provided during training.
Approach: They analyze how language models acquire character-level knowledge by comparing them to standard settings.
Outcome: The results show that LMs do not treat words as opaque tokens, but instead treat them as tokens.
Understanding Pure Character-Based Neural Machine Translation: The Case of Translating Finnish into English (2020.coling-main)

Copied to clipboard

Challenge: Recent work shows that deeper character-based neural machine translation models outperform subword-based models.
Approach: They propose to investigate the ability of character-based models to learn word senses and morphological inflections and the attention mechanism in Finnish into English translation.
Outcome: The character-based models outperform subword-based model in Finnish to English translation.
Pretrained Language Models for Sequential Sentence Classification (D19-1)

Copied to clipboard

Challenge: Recent successful models for document-level understanding have used hierarchical encoding and CRFs to capture dependencies between subsequent labels.
Approach: They propose a pretrained language model that captures contextual dependencies without hierarchical encoding nor a CRF.
Outcome: The proposed model captures contextual dependencies without hierarchical encoding nor a CRF on four datasets, including a new dataset of structured scientific abstracts.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations