Papers by Richard Sproat
Semi-supervised URL Segmentation with Recurrent Neural Networks Pre-trained on Knowledge Graph Entities (2020.coling-main)
Copied to clipboard
| Challenge: | Domain names such as openresearch are being added to a growing set of tokens that an NLP system may need to deal with. |
| Approach: | They propose a tagging model that uses characters as input to break domain names into component words . they propose taagging methods that use concatenated entity names in a large knowledge database . |
| Outcome: | The proposed model improves on concatenated entity names in a knowledge database by 33% . the proposed model can be used for a wide range of languages, including Chinese and Japanese . |
Creating ConLangs to Probe the Metalinguistic Grammatical Knowledge of LLMs (2026.findings-acl)
Copied to clipboard
| Challenge: | 'ConLang' is a term used to describe any artificially created language intended to be as expressive as naturally evolved human languages. |
| Approach: | They propose to use large language models to create a modular system that uses LLMs as a tool in the development of Constructed Languages. |
| Outcome: | The proposed system creates phonology, morphology and syntax, lexicon, orthography, and grammatical handbook using module-specific sets of prompts. |
Structured abbreviation expansion in context (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Ad hoc abbreviations are commonly found in informal communication channels that favor shorter messages. |
| Approach: | They propose to reverse ad hoc abbreviations in context to recover normalized, expanded versions of abbrevated messages. |
| Outcome: | The proposed method can recover normalized, expanded abbreviations from text . it is similar to spelling correction, but requires more extensive work . |
Bi-Phone: Modeling Inter Language Phonetic Influences in Text (2023.acl-long)
Copied to clipboard
Abhirut Gupta, Ananya B. Sai, Richard Sproat, Yuri Vasilevski, James Ren, Ambarish Jash, Sukhdeep Sodhi, Aravindan Raghuveer
| Challenge: | Increasingly, people are forced to use the Web in languages they have low literacy in due to technology asymmetries. |
| Approach: | They propose a method to mine phoneme confusions for pairs of L1 and L2 and plug them into a generative model for synthetically producing corrupted L2 text. |
| Outcome: | The proposed method corrupts the popular language understanding benchmark SuperGLUE and improves performance. |
Fast and Accurate Reordering with ITG Transition RNN (C18-1)
Copied to clipboard
| Challenge: | Attention-based sequence-to-sequence neural networks learn to jointly align and translate. |
| Approach: | They propose to use a reordering RNN that shares the input encoder with the decoder to decouple re-ordering from translation. |
| Outcome: | The proposed model can achieve superior reordering accuracy without feature engineering and is 2.5x faster in decoding. |
Helpful Neighbors: Leveraging Neighbors in Geographic Feature Pronunciation (2023.tacl-1)
Copied to clipboard
| Challenge: | a new architecture learns to use pronunciations of neighboring names to guess pronunciations . features cause not infrequent problems in the US, but become a serious issue in Japan . |
| Approach: | They propose an architecture that learns to use pronunciations of neighboring names to guess pronunciations . they propose corrections for errors in Google Maps and an application to a totally different task . |
| Outcome: | The proposed model can be applied to finding and proposing corrections for errors in Google Maps. |