Papers by Isaac Caswell
Language ID in the Wild: Unexpected Challenges on the Path to a Thousand-Language Web Text Corpus (2020.coling-main)
Copied to clipboard
| Challenge: | Large text corpora are increasingly important for a wide variety of NLP tasks. |
| Approach: | They propose to train automatic language identification models on up to 1,629 languages . they find that human-judged accuracy for web-crawl text corpora is only around 5% for many lower-resource languages. |
| Outcome: | The proposed models achieve over 90% average F1 on 1,629 languages . human-judged accuracy for web-crawl text corpora is only around 5% for many lower-resource languages - suggesting a need for more robust evaluation. |
Learning a Multi-Domain Curriculum for Neural Machine Translation (2020.acl-main)
Copied to clipboard
| Challenge: | Existing data selection methods do not work well for multiple domains . multiple aspects need to be considered for training a multi-domain model . |
| Approach: | They propose a dynamic data selection method to multi-domain NMT that incorporates instance-level domain-relevance features and a curriculum to gradually focus on multi- domain relevant data batches. |
| Outcome: | The proposed model outperforms no-curriculum training on multiple domains and reaches or outperformed individual performance. |
Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets (2022.tacl-1)
Copied to clipboard
Julia Kreutzer, Isaac Caswell, Lisa Wang, Ahsan Wahab, Daan van Esch, Nasanbayar Ulzii-Orshikh, Allahsera Tapo, Nishant Subramani, Artem Sokolov, Claytone Sikasote, Monang Setyawan, Supheakmungkol Sarin, Sokhar Samb, Benoît Sagot, Clara Rivera, Annette Rios, Isabel Papadimitriou, Salomey Osei, Pedro Ortiz Suarez, Iroro Orife, Kelechi Ogueji, Andre Niyongabo Rubungo, Toan Q. Nguyen, Mathias Müller, André Müller, Shamsuddeen Hassan Muhammad, Nanda Muhammad, Ayanda Mnyakeni, Jamshidbek Mirzakhalov, Tapiwanashe Matangira, Colin Leong, Nze Lawson, Sneha Kudugunta, Yacine Jernite, Mathias Jenny, Orhan Firat, Bonaventure F. P. Dossou, Sakhile Dlamini, Nisansa de Silva, Sakine Çabuk Ballı, Stella Biderman, Alessia Battisti, Ahmed Baruwa, Ankur Bapna, Pallavi Baljekar, Israel Abebe Azime, Ayodele Awokoya, Duygu Ataman, Orevaoghene Ahia, Oghenefego Ahia, Sweta Agrawal, Mofetoluwa Adeyemi
| Challenge: | Lower-resource corpora have systematic issues, including mislabeled or nonstandard/ambiguous language codes. |
| Approach: | They manually audit the quality of 205 language-specific corpora released with five major public datasets. |
| Outcome: | The results show that lower-resource corpora have systematic issues even for non-proficient speakers. |
Investigating Multilingual NMT Representations at Scale (D19-1)
Copied to clipboard
| Challenge: | Multilingual Neural Machine Translation models have shown success in transfer learning settings, but their mode of transfer remains elusive. |
| Approach: | They propose to use a representation similarity framework to compare multilingual representations using a SVCCA representation similar to the previous work. |
| Outcome: | The proposed model can be used to compare representations across languages and layers. |
Dynamically Composing Domain-Data Selection with Clean-Data Selection by “Co-Curricular Learning” for Neural Machine Translation (P19-1)
Copied to clipboard
| Challenge: | Existing studies focus separately on domain-data selection, clean-data selecting, or their static combination, leaving the dynamic interaction across them not explicitly examined. |
| Approach: | They propose a method to combine dynamic domain-data selection with dynamic clean-data selecting for transfer learning across both capabilities. |
| Outcome: | The proposed method performs well on two domains and shows the properties of the data scheduled by the co-curriculum. |
Connecting Language Technologies with Rich, Diverse Data Sources Covering Thousands of Languages (2024.lrec-main)
Copied to clipboard
Daan van Esch, Sandy Ritchie, Sebastian Ruder, Julia Kreutzer, Clara Rivera, Ishank Saxena, Isaac Caswell
| Challenge: | Existing data sources for many thousands of languages are rich and diverse . Efforts are ongoing to extend technology to many more of the world's languages . |
| Approach: | They provide an overview of some of the major online data sources available for thousands of languages. |
| Outcome: | The proposed language technologies are based on the data available for thousands of languages. |
XTREME-UP: A User-Centric Scarce-Data Benchmark for Under-Represented Languages (2023.findings-emnlp)
Copied to clipboard
Sebastian Ruder, Jonathan Clark, Alexander Gutkin, Mihir Kale, Min Ma, Massimo Nicosia, Shruti Rijhwani, Parker Riley, Jean-Michel Sarr, Xinyi Wang, John Wieting, Nitish Gupta, Anna Katanova, Christo Kirov, Dana Dickinson, Brian Roark, Bidisha Samanta, Connie Tao, David Adelani, Vera Axelrod, Isaac Caswell, Colin Cherry, Dan Garrette, Reeve Ingle, Melvin Johnson, Dmitry Panteleev, Partha Talukdar
| Challenge: | Existing datasets are often informed by established research directions in the NLP community. |
| Approach: | They propose a benchmark to evaluate the capabilities of language models across 88 under-represented languages over 9 key user-centric technologies including ASR, OCR, MT, and information access tasks. |
| Outcome: | The proposed benchmark evaluates the capabilities of language models across 88 under-represented languages over 9 key user-centric technologies including ASR, OCR, MT, and information access tasks. |
BLEU might be Guilty but References are not Innocent (2020.emnlp-main)
Copied to clipboard
| Challenge: | Using a method to collect references and compare their value with human evaluations, we show that multi-reference BLEU does not improve the correlation for high quality output. |
| Approach: | They propose a method to compare the quality of automated metrics by analyzing references and comparing them with human evaluations. |
| Outcome: | The proposed method improves correlation with all modern evaluation metrics including embedding-based methods. |
Translationese as a Language in “Multilingual” NMT (2020.acl-main)
Copied to clipboard
| Challenge: | Recent work examines the impact of translationese in machine translation evaluation using the WMT evaluation campaign. |
| Approach: | They propose to use a sentence-level classifier to distinguish translationese from original target text to generate a machine translation model that can produce more natural outputs at test time. |
| Outcome: | The proposed model produces more natural outputs at test time, yielding gains in human evaluation scores on accuracy and fluency. |
Writing System and Speaker Metadata for 2,800+ Language Varieties (2022.lrec-1)
Copied to clipboard
| Challenge: | Currently, language technologies are easily available in only a small minority of the world's 7,000+ language varieties. |
| Approach: | They propose to use an open-source dataset to provide the writing system(s) for each of the 2,800+ languages used in the world today and an estimated speaker count for each. |
| Outcome: | The dataset provides the attested writing system(s) for each of these 2,800+ varieties, as well as an estimated speaker count for each variety. |
GATITOS: Using a New Multilingual Lexicon for Low-resource Machine Translation (2023.emnlp-main)
Copied to clipboard
| Challenge: | a new study explores the effectiveness of bilingual lexica in machine translation models . cross-lingual vocabulary alignment is still highly imperfect in these models, despite the success of supervised and self-supervised training. |
| Approach: | They use a resource to improve translation performance on 200-language models . they show that lexica is more reliable than human-translated data . |
| Outcome: | The proposed approach improves on 200-language translation models with lexical data augmentation . the proposed approach is open-source and has 168 tail languages . |