2 |
[The Routledge Course in Modern Mandarin Chinese - Workbook 2 - Simplified Characters Edition] ; Routledge Kurs in moderner chinesischer Hochsprache : Arbeitsbuch 2 (Ausgabe mit Kurzzeichen)
|
|
|
|
DNB Subject Category Language
|
|
Show details
|
|
3 |
[The Routledge Course in Modern Mandarin Chinese - Textbook 2 - Simplified Characters Edition] ; Routledge Kurs in moderner chinesischer Hochsprache : Lehrbuch 2 (Ausgabe mit Kurzzeichen)
|
|
|
|
DNB Subject Category Language
|
|
Show details
|
|
4 |
[The Routledge Course in Modern Mandarin Chinese - Workbook 1 (Simplified Characters)] ; Routledge Kurs in moderner chinesischer Hochsprache : Arbeitsbuch 1 (Ausgabe in Kurzzeichen)
|
|
|
|
DNB Subject Category Language
|
|
Show details
|
|
5 |
[The Routledge Course in Modern Mandarin Chinese - Textbook Level 1 (Simplified Characters)] ; Routledge Kurs in moderner chinesischer Hochsprache : Lehrbuch 1 (Ausgabe mit Kurzzeichen)
|
|
|
|
DNB Subject Category Language
|
|
Show details
|
|
7 |
Documenting Geographically and Contextually Diverse Data Sources: The BigScience Catalogue of Language Data and Resources
|
|
|
|
In: https://hal.inria.fr/hal-03550289 ; 2022 (2022)
|
|
BASE
|
|
Show details
|
|
9 |
Applying Intertextuality to Chinese Traditional Animation Making in a Global Context: Using Jingwei Reclamation as an Example
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Hybrid system and flipped classroom: digital lexiculturology in CFL ; Dispositif hybride et classe inversée : lexiculturologie numérique en CLE
|
|
|
|
In: Journées d’études Enseignement en ligne de la langue et de la civilisation chinoises. ; https://hal.archives-ouvertes.fr/hal-03643992 ; Journées d’études Enseignement en ligne de la langue et de la civilisation chinoises., Jun 2022, Paris, France (2022)
|
|
BASE
|
|
Show details
|
|
11 |
Analyse contrastive de la métalexicographie et de la lexicographie françaises et chinoises
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-03596265 ; 2022 (2022)
|
|
BASE
|
|
Show details
|
|
12 |
Addressing racial/ethnic inequities in vaccine hesitancy and uptake: lessons learned from the California alliance against COVID-19.
|
|
|
|
BASE
|
|
Show details
|
|
13 |
Phraséoculturologie et phraséodidactique en LE : comment intégrer progressivement les expressions idiomatiques dans les manuels à partir des niveaux A1-B2 ?
|
|
|
|
In: Colloque international : Enjeux théoriques et pratiques méthodologiques ; https://hal.archives-ouvertes.fr/hal-03620340 ; Colloque international : Enjeux théoriques et pratiques méthodologiques, Jun 2022, Grenoble, France (2022)
|
|
BASE
|
|
Show details
|
|
14 |
Analyse contrastive sur la pragmatique des expressions idiomatiques françaises et chinoises : les défigements
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-03589947 ; 2022 (2022)
|
|
BASE
|
|
Show details
|
|
15 |
Imputing Out-of-Vocabulary Embeddings with LOVE Makes Language Models Robust with Little Cost
|
|
|
|
In: ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics ; https://hal.archives-ouvertes.fr/hal-03613101 ; ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, May 2022, Dublin, Ireland (2022)
|
|
Abstract:
International audience ; State-of-the-art NLP systems represent inputs with word embeddings, but these are brittle when faced with Out-of-Vocabulary (OOV) words. To address this issue, we follow the principle of mimick-like models to generate vectors for unseen words, by learning the behavior of pre-trained embeddings using only the surface form of words. We present a simple contrastive learning framework, LOVE, which extends the word representation of an existing pre-trained language model (such as BERT), and makes it robust to OOV with few additional parameters. Extensive evaluations demonstrate that our lightweight model achieves similar or even better performances than prior competitors, both on original datasets and on corrupted variants. Moreover, it can be used in a plug-and-play fashion with FastText and BERT, where it significantly improves their robustness.
|
|
Keyword:
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]; Language models; Out-of-vocabulary OOV words; Word embeddings
|
|
URL: https://hal.archives-ouvertes.fr/hal-03613101/file/OOV_Problem.pdf https://hal.archives-ouvertes.fr/hal-03613101 https://hal.archives-ouvertes.fr/hal-03613101/document
|
|
BASE
|
|
Hide details
|
|
16 |
Imputing out-of-vocabulary embeddings with LOVE makes language models robust with little cost
|
|
|
|
In: ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics ; https://hal.archives-ouvertes.fr/hal-03613101 ; ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, May 2022, Dublin, Ireland (2022)
|
|
BASE
|
|
Show details
|
|
17 |
Contrastive phraseocultural analysis : Stereotypes and representations of the head in Chinese and French
|
|
|
|
In: 13th International Conference of Corpus Linguistics (CILC2022) ; https://hal.archives-ouvertes.fr/hal-03596954 ; 13th International Conference of Corpus Linguistics (CILC2022), May 2022, Bergamo, Italy (2022)
|
|
BASE
|
|
Show details
|
|
18 |
Presuppositions in If-Conditionals: Testing for Asymmetry ...
|
|
|
|
BASE
|
|
Show details
|
|
19 |
Generating Authentic Adversarial Examples beyond Meaning-preserving with Doubly Round-trip Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
20 |
Conditional Bilingual Mutual Information Based Adaptive Training for Neural Machine Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|