Publications by OLST members

@inproceedings{santurkar_whose_2023,

title = {Whose Opinions Do Language Models Reflect?},

author = {Shibani Santurkar and Esin Durmus and Faisal Ladhak and Cinoo Lee and Percy Liang and Tatsunori Hashimoto},

year  = {2023},

date = {2023-01-01},

booktitle = {Proceedings of the 40th International Conference on Machine Learning},

pages = {1–34},

publisher = {Proceedings of Machine Learning Research},

address = {Honolulu, Hawaii, USA},

abstract = {Language models (LMs) are increasingly being used in open-ended contexts, where the opinions they reflect in response to subjective queries can have a profound impact, both on user satisfaction, and shaping the views of society at large. We put forth a quantitative framework to investigate the opinions reflected by LMs – by leveraging high-quality public opinion polls. Using this framework, we create OpinionQA, a dataset for evaluating the alignment of LM opinions with those of 60 US demographic groups over topics ranging from abortion to automation. Across topics, we find substantial misalignment between the views reflected by current LMs and those of US demographic groups: on par with the DemocratRepublican divide on climate change. Notably, this misalignment persists even after explicitly steering the LMs towards particular groups. Our analysis not only confirms prior observations about the left-leaning tendencies of some human feedback-tuned LMs, but also surfaces groups whose opinions are poorly reflected by current LMs (e.g., 65+ and widowed individuals).},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{marshman_weird_2022,

title = {Weird, Wonderful, Worthy, and Worrying: Use Cases for MT as Described in Canadian Newspapers},

author = {Elizabeth Marshman},

url = {http://acl-anthology.online/nettt-2022/},

year  = {2022},

date = {2022-07-01},

urldate = {2023-06-06},

booktitle = {Proceedings of the International Conference New Trends in Translation and Technology NeTTT 2022},

pages = {196–205},

address = {Rhodes Island, Greece},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{dube_handling_2022,

title = {Handling Idioms in Symbolic Multilingual Natural Language Generation},

author = {Michaelle Dubé and François Lareau},

editor = {Archna Bhatia and Paul Cook and Shiva Taslimipoor and Marcos Garcia and Carlos Ramisch},

url = {https://aclanthology.org/2022.mwe-1.17},

year  = {2022},

date = {2022-06-01},

urldate = {2024-05-31},

booktitle = {Proceedings of the 18th Workshop on Multiword Expressions @LREC2022},

pages = {118–126},

publisher = {European Language Resources Association},

address = {Marseille, France},

abstract = {While idioms are usually very rigid in their expression, they sometimes allow a certain level of freedom in their usage, with modifiers or complements splitting them or being syntactically attached to internal nodes rather than to the root (e.g., “take something with a big grain of salt”). This means that they cannot always be handled as ready-made strings in rule-based natural language generation systems. Having access to the internal syntactic structure of an idiom allows for more subtle processing. We propose a way to enumerate all possible language-independent n-node trees and to map particular idioms of a language onto these generic syntactic patterns. Using this method, we integrate the idioms from the LN-fr into GenDR, a multilingual realizer. Our implementation covers nearly 98% of LN-fr's idioms with high precision, and can easily be extended or ported to other languages.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{kletz_methodology_2022,

title = {A Methodology for Building a Diachronic Dataset of Semantic Shifts and its Application to QC-FR-Diac-V1.0, a Free Reference for French},

author = {David Kletz and Philippe Langlais and François Lareau and Patrick Drouin},

editor = {Nicoletta Calzolari and Frédéric Béchet and Philippe Blache and Khalid Choukri and Christopher Cieri and Thierry Declerck and Sara Goggi and Hitoshi Isahara and Bente Maegaard and Joseph Mariani and Hélène Mazo and Jan Odijk and Stelios Piperidis},

url = {https://aclanthology.org/2022.lrec-1.228},

year  = {2022},

date = {2022-06-01},

urldate = {2024-05-31},

booktitle = {Proceedings of the Thirteenth Language Resources and Evaluation Conference},

pages = {2117–2125},

publisher = {European Language Resources Association},

address = {Marseille, France},

abstract = {Different algorithms have been proposed to detect semantic shifts (changes in a word meaning over time) in a diachronic corpus. Yet, and somehow surprisingly, no reference corpus has been designed so far to evaluate them, leaving researchers to fallback to troublesome evaluation strategies. In this work, we introduce a methodology for the construction of a reference dataset for the evaluation of semantic shift detection, that is, a list of words where we know for sure whether they present a word meaning change over a period of interest. We leverage a state-of-the-art word-sense disambiguation model to associate a date of first appearance to all the senses of a word. Significant changes in sense distributions as well as clear stability are detected and the resulting words are inspected by experts using a dedicated interface before populating a reference dataset. As a proof of concept, we apply this methodology to a corpus of newspapers from Quebec covering the whole 20th century. We manually verified a subset of candidates, leading to QC-FR-Diac-V1.0, a corpus of 151 words allowing one to evaluate the identification of semantic shifts in French between 1910 and 1990.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{hershcovich_challenges_2022,

title = {Challenges and Strategies in Cross-Cultural NLP},

author = {Daniel Hershcovich and Stella Frank and Heather Lent and Miryam De Lhoneux and Mostafa Abdou and Stephanie Brandl and Emanuele Bugliarello and Laura Cabello Piqueras and Ilias Chalkidis and Ruixiang Cui and Constanza Fierro and Katerina Margatina and Phillip Rust and Anders Søgaard},

url = {https://aclanthology.org/2022.acl-long.482},

doi = {10.18653/v1/2022.acl-long.482},

year  = {2022},

date = {2022-01-01},

urldate = {2024-11-08},

booktitle = {Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},

pages = {6997–7013},

publisher = {Association for Computational Linguistics},

address = {Dublin, Ireland},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{rigouts_terryn_d-terminer_2022,

title = {D-Terminer: Online Demo for Monolingual and Bilingual Automatic Term Extraction},

author = {Ayla Rigouts Terryn and Veronique Hoste and Els Lefever},

year  = {2022},

date = {2022-01-01},

booktitle = {Proceedings of the Workshop on Terminology in the 21st century: many faces, many places @LREC2022},

pages = {33–40},

publisher = {European Language Resources Association},

address = {Marseille, France},

abstract = {This contribution presents D-Terminer: an open access, online demo for monolingual and multilingual automatic term extraction from parallel corpora. The monolingual term extraction is based on a recurrent neural network, with a supervised methodology that relies on pretrained embeddings. Candidate terms can be tagged in their original context and there is no need for a large corpus, as the methodology will work even for single sentences. With the bilingual term extraction from parallel corpora, potentially equivalent candidate term pairs are extracted from translation memories and manual annotation of the results shows that good equivalents are found for most candidate terms. Accompanying the release of the demo is an updated version of the ACTER Annotated Corpora for Term Extraction Research (version 1.5).},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{rigouts_terryn_acter_2022,

title = {ACTER 1.5: Annotated Corpora for Term Extraction Research},

author = {Ayla Rigouts Terryn and Veronique Hoste and Els Lefever},

year  = {2022},

date = {2022-01-01},

booktitle = {CLARIN Annual Conference Proceedings},

pages = {1–4},

address = {Prague, Czech Republic},

abstract = {This contribution presents version 1.5 of the Annotated Corpora for Term Extraction Research (ACTER) dataset. It includes domain-specific corpora in three languages (English, French, and Dutch) and four domains (corruption, dressage (equitation), heart failure, and wind energy). Manual annotations are available of terms and Named Entities for each corpus, with almost 20k unique annotations in total. Significant improvements have been made, most notably the inclusion of sequential annotations. Additionally, an online demo – D-Termine – has been launched for monolingual and bilingual automatic term extraction from parallel corpora, based on the dataset.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{joan_casademont_allers-retours_2022-1,

title = {Allers-retours entre recherche et pratique : Analyse de besoins et capsules complémentaires au service des enseignants de Catalan langue additionnelle},

author = {Joan Casademont, Anna and Gagné, Nancy and Viladrich Castellanas, Èric},

year  = {2022},

date = {2022-01-01},

publisher = {REFAD, ONE, CIRTA, Université TÉLUQ},

address = {Québec},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{shwartz_good_2022,

title = {Good Night at 4 pm?! Time Expressions in Different Cultures},

author = {Vered Shwartz},

url = {https://aclanthology.org/2022.findings-acl.224},

doi = {10.18653/v1/2022.findings-acl.224},

year  = {2022},

date = {2022-01-01},

urldate = {2025-02-20},

booktitle = {Findings of the Association for Computational Linguistics: ACL 2022},

pages = {2842–2853},

publisher = {Association for Computational Linguistics},

address = {Dublin, Ireland},

abstract = {We propose the task of culture-specific time expression grounding, i.e. mapping from expressions such as “morning” in English or “manhã” in Portuguese to specific hours in the day. We propose 3 language-agnostic methods, one of which achieves promising results on gold standard annotations that we collected for a small number of languages. We then apply this method to 27 languages and analyze the similarities across languages in the grounding of time expressions.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{miotto_who_2022,

title = {Who is GPT-3? An exploration of personality, values and demographics},

author = {Marilù Miotto and Nicola Rossberg and Bennett Kleinberg},

url = {https://aclanthology.org/2022.nlpcss-1.24},

doi = {10.18653/v1/2022.nlpcss-1.24},

year  = {2022},

date = {2022-01-01},

urldate = {2024-10-28},

booktitle = {Proceedings of the Fifth Workshop on Natural Language Processing and Computational Social Science (NLP+CSS)},

pages = {218–227},

publisher = {Association for Computational Linguistics},

address = {Abu Dhabi, UAE},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{teichmann_generic_2021,

title = {Generic Oracles for Structured Prediction},

author = {Christoph Teichmann and Antoine Venant},

editor = {Stephan Oepen and Kenji Sagae and Reut Tsarfaty and Gosse Bouma and Djamé Seddah and Daniel Zeman},

url = {https://aclanthology.org/2021.iwpt-1.1},

doi = {10.18653/v1/2021.iwpt-1.1},

year  = {2021},

date = {2021-08-01},

booktitle = {Proceedings of the 17th International Conference on Parsing Technologies and the IWPT 2021 Shared Task on Parsing into Enhanced Universal Dependencies (IWPT 2021)},

pages = {1–12},

publisher = {Association for Computational Linguistics},

address = {Online},

abstract = {When learned without exploration, local models for structured prediction tasks are subject to exposure bias and cannot be trained without detailed guidance. Active Imitation Learning (AIL), also known in NLP as Dynamic Oracle Learning, is a general technique for working around these issues by allowing the exploration of different outputs at training time. AIL requires oracle feedback: an oracle is any algorithm which can, given a partial candidate solution and gold annotation, find the correct (minimum loss) next output to produce. This paper describes a general finite state technique for deriving oracles. The technique describe is also efficient and will greatly expand the tasks for which AIL can be used.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{kassner_multilingual_2021,

title = {Multilingual LAMA: Investigating Knowledge in Multilingual Pretrained Language Models},

author = {Nora Kassner and Philipp Dufter and Hinrich Schütze},

url = {https://aclanthology.org/2021.eacl-main.284},

doi = {10.18653/v1/2021.eacl-main.284},

year  = {2021},

date = {2021-01-01},

urldate = {2025-02-20},

booktitle = {Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume},

pages = {3250–3258},

publisher = {Association for Computational Linguistics},

address = {Online},

abstract = {Recently, it has been found that monolingual English language models can be used as knowledge bases. Instead of structural knowledge base queries, masked sentences such as “Paris is the capital of [MASK]” are used as probes. We translate the established benchmarks TREx and GoogleRE into 53 languages. Working with mBERT, we investigate three questions. (i) Can mBERT be used as a multilingual knowledge base? Most prior work only considers English. Extending research to multiple languages is important for diversity and accessibility. (ii) Is mBERT’s performance as knowledge base language-independent or does it vary from language to language? (iii) A multilingual model is trained on more text, e.g., mBERT is trained on 104 Wikipedias. Can mBERT leverage this for better performance? We find that using mBERT as a knowledge base yields varying performance across languages and pooling predictions across languages improves performance. Conversely, mBERT exhibits a language bias; e.g., when queried in Italian, it tends to predict Italy as the country of origin.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}