paper-with-me

홈 › Papers

Models In a Spelling Bee: Language Models Implicitly Learn the Character Composition of Tokens

2021-08-25 · NAACL 2022 7 · Itay Itzhak, Omer Levy

Standard pretrained language models operate on sequences of subword tokens without direct access to the characters that compose each token's string representation. We probe the embedding layer of pretrained language models and show that models learn the internal character composition of whole word and subword tokens to a surprising extent, without ever seeing the characters coupled with the tokens. Our results show that the embedding layer of RoBERTa holds enough information to accurately spell up to a third of the vocabulary and reach high average character ngram overlap on all token types. We further test whether enriching subword models with additional character information can improve language modeling, and observe that this method has a near-identical learning curve as training without spelling-based enrichment. Overall, our results suggest that language modeling objectives incentivize the model to implicitly learn some notion of spelling, and that explicitly teaching the model how to spell does not appear to enhance its performance on such tasks.

📄 PDF Abstract BibTeX arXiv:2108.11193

Code (1)

itay1itzhak/spellingbee 공식 구현 pytorch

Tasks

Language ModelingLanguage Modelling

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Weight Decay 설명 없음
Residual Connection 설명 없음
Multi-Head Attention 설명 없음

Similar Papers 제목 키워드 기반

Models In a Spelling Bee: Language Models Implicitly Learn the Character Composition of Tokens

2021-10-16 · ACL ARR October 2021 10 · Anonymous

Standard pretrained language models operate on sequences of subword tokens without direct access to the characters that compose each token’s string representation. We probe the embedding layer of pretrained language mode…

Language ModelingLanguage Modelling

Spelling-out is not Straightforward: LLMs' Capability of Tokenization from Token to Characters

2025-06-12 · Tatsuya Hiraoka, Kentaro Inui

Large language models (LLMs) can spell out tokens character by character with high accuracy, yet they struggle with more complex character-level tasks, such as identifying compositional subcomponents within tokens. In th…

Char2Subword: Extending the Subword Embedding Space Using Robust Character Compositionality

2020-10-24 · Findings (EMNLP) 2021 11 · Gustavo Aguilar, Bryan McCann, Tong Niu, Nazneen Rajani 외

Byte-pair encoding (BPE) is a ubiquitous algorithm in the subword tokenization process of language models as it provides multiple benefits. However, this process is solely based on pre-training data statistics, making it…

HandReader: Advanced Techniques for Efficient Fingerspelling Recognition

2025-05-15 · Pavel Korotaev, Petr Surovtsev, Alexander Kapitanov, Karina Kvanchiani 외

Fingerspelling is a significant component of Sign Language (SL), allowing the interpretation of proper names, characterized by fast hand movements during signing. Although previous works on fingerspelling recognition hav…

Sign Language Recognition

Fingerspelling within Sign Language Translation

2024-08-13 · Garrett Tanzer

Fingerspelling poses challenges for sign language processing due to its high-frequency motion and use for open-vocabulary terms. While prior work has studied fingerspelling recognition, there has been little attention to…

Sign Language TranslationTranslation