Papers by Keith Vertanen
Accelerating Text Communication via Abbreviated Sentence Input (2021.acl-long)
Copied to clipboard
| Challenge: | Skipping spaces or other characters may be able to speed input and reduce a user’s physical input effort. |
| Approach: | They designed a neural language model to expand noisy abbreviated input where users often omit spaces and mid-word vowels. |
| Outcome: | The proposed recognizer can expand noisy abbreviated input even if a third of characters is omitted. |
Adapting Large Language Models for Character-based Augmentative and Alternative Communication (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Most character language models predict subword tokens of variable length . |
| Approach: | They propose to use large pretrained character language models to make accurate character predictions. |
| Outcome: | The proposed method produces more accurate character predictions than classification models and n-gram models. |