Search

Your search keyword '"Fourrier, Clémentine"' showing total 14 results
Start Over Author "Fourrier, Clémentine"
14 results on '"Fourrier, Clémentine"'

1. The Hallucinations Leaderboard -- An Open Effort to Measure Hallucinations in Large Language Models

Author

Hong, Giwon, Gema, Aryo Pradipta, Saxena, Rohit, Du, Xiaotang, Nie, Ping, Zhao, Yu, Perez-Beltrachini, Laura, Ryabinin, Max, He, Xuanli, Fourrier, Clémentine, and Minervini, Pasquale

Subjects
Computer Science - Computation and Language
Abstract

Large Language Models (LLMs) have transformed the Natural Language Processing (NLP) landscape with their remarkable ability to understand and generate human-like text. However, these models are prone to ``hallucinations'' -- outputs that do not align with factual reality or the input context. This paper introduces the Hallucinations Leaderboard, an open initiative to quantitatively measure and compare the tendency of each model to produce hallucinations. The leaderboard uses a comprehensive set of benchmarks focusing on different aspects of hallucinations, such as factuality and faithfulness, across various tasks, including question-answering, summarisation, and reading comprehension. Our analysis provides insights into the performance of different models, guiding researchers and practitioners in choosing the most reliable models for their applications.

Published
2024

2. GAIA: a benchmark for General AI Assistants

Author

Mialon, Grégoire, Fourrier, Clémentine, Swift, Craig, Wolf, Thomas, LeCun, Yann, and Scialom, Thomas

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

We introduce GAIA, a benchmark for General AI Assistants that, if solved, would represent a milestone in AI research. GAIA proposes real-world questions that require a set of fundamental abilities such as reasoning, multi-modality handling, web browsing, and generally tool-use proficiency. GAIA questions are conceptually simple for humans yet challenging for most advanced AIs: we show that human respondents obtain 92\% vs. 15\% for GPT-4 equipped with plugins. This notable performance disparity contrasts with the recent trend of LLMs outperforming humans on tasks requiring professional skills in e.g. law or chemistry. GAIA's philosophy departs from the current trend in AI benchmarks suggesting to target tasks that are ever more difficult for humans. We posit that the advent of Artificial General Intelligence (AGI) hinges on a system's capability to exhibit similar robustness as the average human does on such questions. Using GAIA's methodology, we devise 466 questions and their answer. We release our questions while retaining answers to 300 of them to power a leader-board available at https://huggingface.co/gaia-benchmark.

Published
2023

3. Zephyr: Direct Distillation of LM Alignment

Author

Tunstall, Lewis, Beeching, Edward, Lambert, Nathan, Rajani, Nazneen, Rasul, Kashif, Belkada, Younes, Huang, Shengyi, von Werra, Leandro, Fourrier, Clémentine, Habib, Nathan, Sarrazin, Nathan, Sanseviero, Omar, Rush, Alexander M., and Wolf, Thomas

Subjects
Computer Science - Machine Learning, Computer Science - Computation and Language
Abstract

We aim to produce a smaller language model that is aligned to user intent. Previous research has shown that applying distilled supervised fine-tuning (dSFT) on larger models significantly improves task accuracy; however, these models are unaligned, i.e. they do not respond well to natural prompts. To distill this property, we experiment with the use of preference data from AI Feedback (AIF). Starting from a dataset of outputs ranked by a teacher model, we apply distilled direct preference optimization (dDPO) to learn a chat model with significantly improved intent alignment. The approach requires only a few hours of training without any additional sampling during fine-tuning. The final result, Zephyr-7B, sets the state-of-the-art on chat benchmarks for 7B parameter models, and requires no human annotation. In particular, results on MT-Bench show that Zephyr-7B surpasses Llama2-Chat-70B, the best open-access RLHF-based model. Code, models, data, and tutorials for the system are available at https://github.com/huggingface/alignment-handbook.

Published
2023

4. BLOOM: A 176B-Parameter Open-Access Multilingual Language Model

Author

Workshop, BigScience, Scao, Teven Le, Fan, Angela, Akiki, Christopher, Pavlick, Ellie, Ilić, Suzana, Hesslow, Daniel, Castagné, Roman, Luccioni, Alexandra Sasha, Yvon, François, Gallé, Matthias, Tow, Jonathan, Rush, Alexander M., Biderman, Stella, Webson, Albert, Ammanamanchi, Pawan Sasanka, Wang, Thomas, Sagot, Benoît, Muennighoff, Niklas, del Moral, Albert Villanova, Ruwase, Olatunji, Bawden, Rachel, Bekman, Stas, McMillan-Major, Angelina, Beltagy, Iz, Nguyen, Huu, Saulnier, Lucile, Tan, Samson, Suarez, Pedro Ortiz, Sanh, Victor, Laurençon, Hugo, Jernite, Yacine, Launay, Julien, Mitchell, Margaret, Raffel, Colin, Gokaslan, Aaron, Simhi, Adi, Soroa, Aitor, Aji, Alham Fikri, Alfassy, Amit, Rogers, Anna, Nitzav, Ariel Kreisberg, Xu, Canwen, Mou, Chenghao, Emezue, Chris, Klamm, Christopher, Leong, Colin, van Strien, Daniel, Adelani, David Ifeoluwa, Radev, Dragomir, Ponferrada, Eduardo González, Levkovizh, Efrat, Kim, Ethan, Natan, Eyal Bar, De Toni, Francesco, Dupont, Gérard, Kruszewski, Germán, Pistilli, Giada, Elsahar, Hady, Benyamina, Hamza, Tran, Hieu, Yu, Ian, Abdulmumin, Idris, Johnson, Isaac, Gonzalez-Dios, Itziar, de la Rosa, Javier, Chim, Jenny, Dodge, Jesse, Zhu, Jian, Chang, Jonathan, Frohberg, Jörg, Tobing, Joseph, Bhattacharjee, Joydeep, Almubarak, Khalid, Chen, Kimbo, Lo, Kyle, Von Werra, Leandro, Weber, Leon, Phan, Long, allal, Loubna Ben, Tanguy, Ludovic, Dey, Manan, Muñoz, Manuel Romero, Masoud, Maraim, Grandury, María, Šaško, Mario, Huang, Max, Coavoux, Maximin, Singh, Mayank, Jiang, Mike Tian-Jian, Vu, Minh Chien, Jauhar, Mohammad A., Ghaleb, Mustafa, Subramani, Nishant, Kassner, Nora, Khamis, Nurulaqilla, Nguyen, Olivier, Espejel, Omar, de Gibert, Ona, Villegas, Paulo, Henderson, Peter, Colombo, Pierre, Amuok, Priscilla, Lhoest, Quentin, Harliman, Rheza, Bommasani, Rishi, López, Roberto Luis, Ribeiro, Rui, Osei, Salomey, Pyysalo, Sampo, Nagel, Sebastian, Bose, Shamik, Muhammad, Shamsuddeen Hassan, Sharma, Shanya, Longpre, Shayne, Nikpoor, Somaieh, Silberberg, Stanislav, Pai, Suhas, Zink, Sydney, Torrent, Tiago Timponi, Schick, Timo, Thrush, Tristan, Danchev, Valentin, Nikoulina, Vassilina, Laippala, Veronika, Lepercq, Violette, Prabhu, Vrinda, Alyafeai, Zaid, Talat, Zeerak, Raja, Arun, Heinzerling, Benjamin, Si, Chenglei, Taşar, Davut Emre, Salesky, Elizabeth, Mielke, Sabrina J., Lee, Wilson Y., Sharma, Abheesht, Santilli, Andrea, Chaffin, Antoine, Stiegler, Arnaud, Datta, Debajyoti, Szczechla, Eliza, Chhablani, Gunjan, Wang, Han, Pandey, Harshit, Strobelt, Hendrik, Fries, Jason Alan, Rozen, Jos, Gao, Leo, Sutawika, Lintang, Bari, M Saiful, Al-shaibani, Maged S., Manica, Matteo, Nayak, Nihal, Teehan, Ryan, Albanie, Samuel, Shen, Sheng, Ben-David, Srulik, Bach, Stephen H., Kim, Taewoon, Bers, Tali, Fevry, Thibault, Neeraj, Trishala, Thakker, Urmish, Raunak, Vikas, Tang, Xiangru, Yong, Zheng-Xin, Sun, Zhiqing, Brody, Shaked, Uri, Yallow, Tojarieh, Hadar, Roberts, Adam, Chung, Hyung Won, Tae, Jaesung, Phang, Jason, Press, Ofir, Li, Conglong, Narayanan, Deepak, Bourfoune, Hatim, Casper, Jared, Rasley, Jeff, Ryabinin, Max, Mishra, Mayank, Zhang, Minjia, Shoeybi, Mohammad, Peyrounette, Myriam, Patry, Nicolas, Tazi, Nouamane, Sanseviero, Omar, von Platen, Patrick, Cornette, Pierre, Lavallée, Pierre François, Lacroix, Rémi, Rajbhandari, Samyam, Gandhi, Sanchit, Smith, Shaden, Requena, Stéphane, Patil, Suraj, Dettmers, Tim, Baruwa, Ahmed, Singh, Amanpreet, Cheveleva, Anastasia, Ligozat, Anne-Laure, Subramonian, Arjun, Névéol, Aurélie, Lovering, Charles, Garrette, Dan, Tunuguntla, Deepak, Reiter, Ehud, Taktasheva, Ekaterina, Voloshina, Ekaterina, Bogdanov, Eli, Winata, Genta Indra, Schoelkopf, Hailey, Kalo, Jan-Christoph, Novikova, Jekaterina, Forde, Jessica Zosa, Clive, Jordan, Kasai, Jungo, Kawamura, Ken, Hazan, Liam, Carpuat, Marine, Clinciu, Miruna, Kim, Najoung, Cheng, Newton, Serikov, Oleg, Antverg, Omer, van der Wal, Oskar, Zhang, Rui, Zhang, Ruochen, Gehrmann, Sebastian, Mirkin, Shachar, Pais, Shani, Shavrina, Tatiana, Scialom, Thomas, Yun, Tian, Limisiewicz, Tomasz, Rieser, Verena, Protasov, Vitaly, Mikhailov, Vladislav, Pruksachatkun, Yada, Belinkov, Yonatan, Bamberger, Zachary, Kasner, Zdeněk, Rueda, Alice, Pestana, Amanda, Feizpour, Amir, Khan, Ammar, Faranak, Amy, Santos, Ana, Hevia, Anthony, Unldreaj, Antigona, Aghagol, Arash, Abdollahi, Arezoo, Tammour, Aycha, HajiHosseini, Azadeh, Behroozi, Bahareh, Ajibade, Benjamin, Saxena, Bharat, Ferrandis, Carlos Muñoz, McDuff, Daniel, Contractor, Danish, Lansky, David, David, Davis, Kiela, Douwe, Nguyen, Duong A., Tan, Edward, Baylor, Emi, Ozoani, Ezinwanne, Mirza, Fatima, Ononiwu, Frankline, Rezanejad, Habib, Jones, Hessie, Bhattacharya, Indrani, Solaiman, Irene, Sedenko, Irina, Nejadgholi, Isar, Passmore, Jesse, Seltzer, Josh, Sanz, Julio Bonis, Dutra, Livia, Samagaio, Mairon, Elbadri, Maraim, Mieskes, Margot, Gerchick, Marissa, Akinlolu, Martha, McKenna, Michael, Qiu, Mike, Ghauri, Muhammed, Burynok, Mykola, Abrar, Nafis, Rajani, Nazneen, Elkott, Nour, Fahmy, Nour, Samuel, Olanrewaju, An, Ran, Kromann, Rasmus, Hao, Ryan, Alizadeh, Samira, Shubber, Sarmad, Wang, Silas, Roy, Sourav, Viguier, Sylvain, Le, Thanh, Oyebade, Tobi, Le, Trieu, Yang, Yoyo, Nguyen, Zach, Kashyap, Abhinav Ramesh, Palasciano, Alfredo, Callahan, Alison, Shukla, Anima, Miranda-Escalada, Antonio, Singh, Ayush, Beilharz, Benjamin, Wang, Bo, Brito, Caio, Zhou, Chenxi, Jain, Chirag, Xu, Chuxin, Fourrier, Clémentine, Periñán, Daniel León, Molano, Daniel, Yu, Dian, Manjavacas, Enrique, Barth, Fabio, Fuhrimann, Florian, Altay, Gabriel, Bayrak, Giyaseddin, Burns, Gully, Vrabec, Helena U., Bello, Imane, Dash, Ishani, Kang, Jihyun, Giorgi, John, Golde, Jonas, Posada, Jose David, Sivaraman, Karthik Rangasai, Bulchandani, Lokesh, Liu, Lu, Shinzato, Luisa, de Bykhovetz, Madeleine Hahn, Takeuchi, Maiko, Pàmies, Marc, Castillo, Maria A, Nezhurina, Marianna, Sänger, Mario, Samwald, Matthias, Cullan, Michael, Weinberg, Michael, De Wolf, Michiel, Mihaljcic, Mina, Liu, Minna, Freidank, Moritz, Kang, Myungsun, Seelam, Natasha, Dahlberg, Nathan, Broad, Nicholas Michio, Muellner, Nikolaus, Fung, Pascale, Haller, Patrick, Chandrasekhar, Ramya, Eisenberg, Renata, Martin, Robert, Canalli, Rodrigo, Su, Rosaline, Su, Ruisi, Cahyawijaya, Samuel, Garda, Samuele, Deshmukh, Shlok S, Mishra, Shubhanshu, Kiblawi, Sid, Ott, Simon, Sang-aroonsiri, Sinee, Kumar, Srishti, Schweter, Stefan, Bharati, Sushil, Laud, Tanmay, Gigant, Théo, Kainuma, Tomoya, Kusa, Wojciech, Labrak, Yanis, Bajaj, Yash Shailesh, Venkatraman, Yash, Xu, Yifan, Xu, Yingxin, Xu, Yu, Tan, Zhe, Xie, Zhongli, Ye, Zifan, Bras, Mathilde, Belkada, Younes, and Wolf, Thomas

Subjects
Computer Science - Computation and Language
Abstract

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to widespread adoption, most LLMs are developed by resource-rich organizations and are frequently kept from the public. As a step towards democratizing this powerful technology, we present BLOOM, a 176B-parameter open-access language model designed and built thanks to a collaboration of hundreds of researchers. BLOOM is a decoder-only Transformer language model that was trained on the ROOTS corpus, a dataset comprising hundreds of sources in 46 natural and 13 programming languages (59 in total). We find that BLOOM achieves competitive performance on a wide variety of benchmarks, with stronger results after undergoing multitask prompted finetuning. To facilitate future research and applications using LLMs, we publicly release our models and code under the Responsible AI License.

Published
2022

5. Entities, Dates, and Languages: Zero-Shot on Historical Texts with T0

Author

De Toni, Francesco, Akiki, Christopher, de la Rosa, Javier, Fourrier, Clémentine, Manjavacas, Enrique, Schweter, Stefan, and van Strien, Daniel

Subjects
Computer Science - Computation and Language
Abstract

In this work, we explore whether the recently demonstrated zero-shot abilities of the T0 model extend to Named Entity Recognition for out-of-distribution languages and time periods. Using a historical newspaper corpus in 3 languages as test-bed, we use prompts to extract possible named entities. Our results show that a naive approach for prompt-based zero-shot multilingual Named Entity Recognition is error-prone, but highlights the potential of such an approach for historical languages lacking labeled datasets. Moreover, we also find that T0-like models can be probed to predict the publication date and language of a document, which could be very relevant for the study of historical texts.

Published
2022

6. The Alzheimer's Disease Prediction Of Longitudinal Evolution (TADPOLE) Challenge: Results after 1 Year Follow-up

Author

Marinescu, Razvan V., Oxtoby, Neil P., Young, Alexandra L., Bron, Esther E., Toga, Arthur W., Weiner, Michael W., Barkhof, Frederik, Fox, Nick C., Eshaghi, Arman, Toni, Tina, Salaterski, Marcin, Lunina, Veronika, Ansart, Manon, Durrleman, Stanley, Lu, Pascal, Iddi, Samuel, Li, Dan, Thompson, Wesley K., Donohue, Michael C., Nahon, Aviv, Levy, Yarden, Halbersberg, Dan, Cohen, Mariya, Liao, Huiling, Li, Tengfei, Yu, Kaixian, Zhu, Hongtu, Tamez-Pena, Jose G., Ismail, Aya, Wood, Timothy, Bravo, Hector Corrada, Nguyen, Minh, Sun, Nanbo, Feng, Jiashi, Yeo, B. T. Thomas, Chen, Gang, Qi, Ke, Chen, Shiyang, Qiu, Deqiang, Buciuman, Ionut, Kelner, Alex, Pop, Raluca, Rimocea, Denisa, Ghazi, Mostafa M., Nielsen, Mads, Ourselin, Sebastien, Sorensen, Lauge, Venkatraghavan, Vikram, Liu, Keli, Rabe, Christina, Manser, Paul, Hill, Steven M., Howlett, James, Huang, Zhiyue, Kiddle, Steven, Mukherjee, Sach, Rouanet, Anais, Taschler, Bernd, Tom, Brian D. M., White, Simon R., Faux, Noel, Sedai, Suman, Oriol, Javier de Velasco, Clemente, Edgar E. V., Estrada, Karol, Aksman, Leon, Altmann, Andre, Stonnington, Cynthia M., Wang, Yalin, Wu, Jianfeng, Devadas, Vivek, Fourrier, Clementine, Raket, Lars Lau, Sotiras, Aristeidis, Erus, Guray, Doshi, Jimit, Davatzikos, Christos, Vogel, Jacob, Doyle, Andrew, Tam, Angela, Diaz-Papkovich, Alex, Jammeh, Emmanuel, Koval, Igor, Moore, Paul, Lyons, Terry J., Gallacher, John, Tohka, Jussi, Ciszek, Robert, Jedynak, Bruno, Pandya, Kruti, Bilgel, Murat, Engels, William, Cole, Joseph, Golland, Polina, Klein, Stefan, and Alexander, Daniel C.

Subjects
Quantitative Biology - Populations and Evolution, Statistics - Applications
Abstract

We present the findings of "The Alzheimer's Disease Prediction Of Longitudinal Evolution" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of biomarkers. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials., Comment: Presents final results of the TADPOLE competition. 60 pages, 7 tables, 14 figures

Published
2020

7. Approches Neuronales pour la Reconstruction de Mots Historiques

Author

Fourrier, Clémentine, Automatic Language Modelling and ANAlysis & Computational Humanities (ALMAnaCH), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), École Pratique des Hautes Études (EPHE), Université Paris sciences et lettres (PSL), Université PSL (Paris Sciences & Lettres), Benoît Sagot, Laurent Romary, Rachel Bawden, and Fourrier, Clémentine

Subjects
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], Etymologie computationelle, Traitement Automatique des Langues (TAL), Historical words reconstruction, Computational linguistics, Natural Language Processing (NLP), Traduction automatique neuronale, [SHS.LANGUE] Humanities and Social Sciences/Linguistics, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Cognates, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], [INFO.INFO-CL] Computer Science [cs]/Computation and Language [cs.CL], Neural Machine Translation (NMT), Cognats, Linguistique computationnelle, Re-construction de mots historiques, Computational etymology, [SHS.LANGUE]Humanities and Social Sciences/Linguistics
Abstract

In historical linguistics, cognates are words that descend in direct line from a common ancestor, called their proto-form, andtherefore are representative of their respective languages evolutions through time, as well as of the relations between theselanguages synchronically. As they reflect the phonetic history of the languages they belong to, they allow linguists to betterdetermine all manners of synchronic and diachronic linguistic relations (etymology, phylogeny, sound correspondences).Cognates of related languages tend to be linked through systematic phonetic correspondence patterns, which neuralnetworks could well learn to model, being especially good at learning latent patterns. In this dissertation, we seek tomethodically study the applicability of machine translation inspired neural networks to historical word prediction, relyingon the surface similarity of both tasks. We first create an artificial dataset inspired by the phonetic and phonotactic rules ofRomance languages, which allow us to vary task complexity and data size in a controlled environment, therefore identifyingif and under which conditions neural networks were applicable. We then extend our work to real datasets (after havingupdated an etymological database to gather a correct amount of data), study the transferability of our conclusions toreal data, then the applicability of a number of data augmentation techniques to the task, to try to mitigate low-resourcesituations. We finally investigat in more detail our best models, multilingual neural networks. We first confirm that, onthe surface, they seem to capture language relatedness information and phonetic similarity, confirming prior work. Wethen discover, by probing them, that the information they store is actually more complex: our multilingual models actuallyencode a phonetic language model, and learn enough latent historical information to allow decoders to reconstruct the(unseen) proto-form of the studied languages as well or better than bilingual models trained specifically on the task. Thislatent information is likely the explanation for the success of multilingual methods in the previous works, En linguistique historique, les cognats sont des mots qui descendent en ligne directe d'un ancêtre commun, leur proto-forme, et qui sont ainsi représentatifs de l'évolution de leurs langues respectives à travers le temps. Comme ils portent eneux l'histoire phonétique des langues auxquelles ils appartiennent, ils permettent aux linguistes de mieux déterminer toutessortes de relations linguistiques synchroniques et diachroniques (étymologie, phylogénie, correspondances phonétiques).Les cognats de langues apparentées sont liés par des correspondances phonétiques systématiques. Les réseaux deneurones, particulièrement adaptés à l'apprentissage de motifs latents, semblent donc bien un bon outil pour modéliserces correspondances. Dans cette thèse, nous cherchons donc à étudier méthodiquement l'applicabilité de réseaux deneurones spécifiques (inspirés de la traduction automatique) à la `prédiction de mots historiques', en nous appuyantsur les similitudes entre ces deux tâches. Nous créons tout d'abord un jeu de données artificiel à partir des règlesphonétiques et phonotactiques des langues romanes, que nous utilisons pour étudier l'utilisation de nos réseaux ensituation controlée, et identifions ainsi sous quelles conditions les réseaux de neurones sont applicables à notre tâched'intérêt. Nous étendons ensuite notre travail à des données réelles (après avoir mis à jour une base étymologiquespour obtenir d'avantage de données), étudions si nos conclusions précédentes leur sont applicables, puis s'il est possibled'utiliser des techniques d'augmentation des données pour pallier aux manque de ressources de certaines situations.Enfin, nous analysons plus en détail nos meilleurs modèles, les réseaux neuronaux multilingues. Nous confirmons àpartir de leurs résultats bruts qu'ils semblent capturer des informations de parenté linguistique et de similarité phonétique,ce qui confirme des travaux antérieurs. Nous découvrons ensuite en les sondant (probing) que les informations qu'ilsstockent sont en fait plus complexes : nos modèles multilingues encodent en fait un modèle phonétique de la langue, etapprennent suffisamment d'informations diachroniques latentes pour permettre à des décodeurs de reconstruire la proto-forme (non vue) des langues étudiées aussi bien, voire mieux, que des modèles bilingues entraînés spécifiquement surcette tâche. Ces informations latentes expliquent probablement le succès des méthodes multilingues dans les travauxprécédents.

Published
2022

8. Caveats of Measuring Semantic Change of Cognates and Borrowings using Multilingual Word Embeddings

Author

Fourrier, Clémentine, Montariol, Syrielle, Fourrier, Clémentine, Automatic Language Modelling and ANAlysis & Computational Humanities (ALMAnaCH), Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)

Subjects
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], [INFO.INFO-CL] Computer Science [cs]/Computation and Language [cs.CL], [INFO.INFO-DB] Computer Science [cs]/Databases [cs.DB], [SHS.LANGUE]Humanities and Social Sciences/Linguistics, [SHS.LANGUE] Humanities and Social Sciences/Linguistics, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract

International audience; Cognates and borrowings carry different aspects of etymological evolution. In this work, we study semantic change of such items using multilingual word embeddings, both static and contextualised. We underline caveats identified while building and evaluating these embeddings. We release both said embeddings and a newly-built historical words lexicon, containing typed relations between words of varied Romance languages.

Published
2022

11. Sound change and neural networks: preliminary experiments

Author

Fourrier, Clémentine, Automatic Language Modelling and ANAlysis & Computational Humanities (ALMAnaCH), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Benzitoun, Christophe, Braud, Chloé, Huber, Laurine, Langlois, David, Ouni, Slim, Pogodalla, Sylvain, and Schneider, Stéphane

Subjects
cognate predict, [INFO.INFO-CL] Computer Science [cs]/Computation and Language [cs.CL], Prédiction de cognats, Changements phonétiques, Réseaux de neurones, historical ling, Regular sound c, Linguistique historique, neural networks, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]
Abstract

Cognate prediction is a key task in historical linguistics that presents a number of similarities withmachine translation. However, although neural methods are now widespread in machine translation,they are still largely unused in historical linguistics. In this paper, we study the performance ofneural methods (more specifically encoder-decoder networks) for the task of cognate prediction. Wefocus in particular on the types of data that can be used for this task, and compare the performanceof statistical and neural methods. We show that sound correspondances can only be learned usingcognate datasets, and that statistical and neural methods seem to have complementary strengths andweaknesses regarding what they learn about the data., La prédiction de cognats est une tâche clef de la linguistique historique et présente de nombreuses similitudes avec les tâches de traduction automatique. Cependant, alors que cette seconde discipline a vu fleurir l’utilisation de méthodes neuronales, celles-ci restent largement absentes des outils utilisés en linguistique historique. Dans ce papier, nous étudions donc la performance des méthodes neuronales utilisées en traduction (les réseaux encodeur-décodeur) pour la tâche de prédiction de cognats. Nous nous intéressons notamment aux types de données utilisables pour cet apprentissage et comparons les résultats obtenus, sur différents types de données, entre des méthodes statistiques et des méthodes neuronales. Nous montrons que l’apprentissage de correspondances phonétiques n’est possible que sur des paires de cognats, et que les méthodes statistiques et neuronales semblent avoir des forces et faiblesses complémentaires quant à ce qu’elles apprennent des données.

Published
2020

12. Evolution phonologique des langues et réseaux de neurones : travaux préliminaires

Author

Fourrier, Clémentine, Automatic Language Modelling and ANAlysis & Computational Humanities (ALMAnaCH), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Benzitoun, Christophe, Braud, Chloé, Huber, Laurine, Langlois, David, Ouni, Slim, Pogodalla, Sylvain, and Schneider, Stéphane

Subjects
réseaux de neurones, prédiction de cognats, Changements phonétiques, linguistique historique, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]
Abstract

La prédiction de cognats est une tâche clef de la linguistique historique et présente de nombreuses similitudes avec les tâches de traduction automatique. Cependant, alors que cette seconde discipline a vu fleurir l’utilisation de méthodes neuronales, celles-ci restent largement absentes des outils utilisés en linguistique historique. Dans ce papier, nous étudions donc la performance des méthodes neuronales utilisées en traduction (les réseaux encodeur-décodeur) pour la tâche de prédiction de cognats. Nous nous intéressons notamment aux types de données utilisables pour cet apprentissage et comparons les résultats obtenus, sur différents types de données, entre des méthodes statistiques et des méthodes neuronales. Nous montrons que l’apprentissage de correspondances phonétiques n’est possible que sur des paires de cognats, et que les méthodes statistiques et neuronales semblent avoir des forces et faiblesses complémentaires quant à ce qu’elles apprennent des données.

Published
2020

14. Building disease progression models from longitudinal biomarkers

Author

Fourrier, Clémentine, Koval, Igor, Bottani, Simona, Colliot, Olivier, Durrleman, Stanley, Algorithms, models and methods for images and signals of the human brain (ARAMIS), Sorbonne Université (SU)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut du Cerveau et de la Moëlle Epinière = Brain and Spine Institute (ICM), Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Fourrier, Clémentine, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut du Cerveau = Paris Brain Institute (ICM), Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Sorbonne Université (SU)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], and Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)

Subjects
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], [SCCO.COMP] Cognitive science/Computer science, [SCCO.COMP]Cognitive science/Computer science, [SDV.NEU]Life Sciences [q-bio]/Neurons and Cognition [q-bio.NC], [SDV.NEU] Life Sciences [q-bio]/Neurons and Cognition [q-bio.NC], ComputingMilieux_MISCELLANEOUS, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract

International audience

Published
2017

Catalog

Books, media, physical & digital resources
See catalog results