DE eng

Search in the Catalogues and Directories

Hits 1 – 18 of 18

1
SCiL 2022 Editors' Note
In: Proceedings of the Society for Computation in Linguistics (2022)
BASE
Show details
2
Sorting through the noise: Testing robustness of information processing in pre-trained language models ...
BASE
Show details
3
On the Interplay Between Fine-tuning and Composition in Transformers ...
Yu, Lang; Ettinger, Allyson. - : arXiv, 2021
BASE
Show details
4
On the Interplay Between Fine-tuning and Composition in Transformers ...
BASE
Show details
5
Pragmatic competence of pre-trained language models through the lens of discourse connectives ...
BASE
Show details
6
Pragmatic competence of pre-trained language models through the lens of discourse connectives ...
Abstract: As pre-trained language models (LMs) continue to dominate NLP, it is increasingly important that we understand the depth of language capabilities in these models. In this paper, we target pre-trained LMs' competence in pragmatics, with a focus on pragmatics relating to discourse connectives. We formulate cloze-style tests using a combination of naturally-occurring data and controlled inputs drawn from psycholinguistics. We focus on testing models' ability to use pragmatic cues to predict discourse connectives, models' ability to understand implicatures relating to connectives, and the extent to which models show humanlike preferences regarding temporal dynamics of connectives. We find that although models predict connectives reasonably well in the context of naturally-occurring data, when we control contexts to isolate high-level pragmatic cues, model sensitivity is much lower. Models also do not show substantial humanlike temporal preferences. Overall, the findings suggest that at present, dominant ...
Keyword: Computational Linguistics; Language Models; Machine Learning; Machine Learning and Data Mining; Natural Language Processing
URL: https://underline.io/lecture/39856-pragmatic-competence-of-pre-trained-language-models-through-the-lens-of-discourse-connectives
https://dx.doi.org/10.48448/x840-9k05
BASE
Hide details
7
Preface: SCiL 2021 Editors' Note
In: Proceedings of the Society for Computation in Linguistics (2021)
BASE
Show details
8
Exploring BERT's Sensitivity to Lexical Cues using Tests from Semantic Priming ...
BASE
Show details
9
Assessing Phrasal Representation and Composition in Transformers ...
Yu, Lang; Ettinger, Allyson. - : arXiv, 2020
BASE
Show details
10
Preface: SCiL 2020 Editors' Note
In: Proceedings of the Society for Computation in Linguistics (2020)
BASE
Show details
11
What BERT Is Not: Lessons from a New Suite of Psycholinguistic Diagnostics for Language Models
In: Transactions of the Association for Computational Linguistics, Vol 8, Pp 34-48 (2020) (2020)
BASE
Show details
12
Mandarin utterance-final particle ba (吧) in the conversational scoreboard
In: Sinn und Bedeutung; Bd. 19 (2015): Proceedings of Sinn und Bedeutung 19; 232-251 ; Proceedings of Sinn und Bedeutung; Vol 19 (2015): Proceedings of Sinn und Bedeutung 19; 232-251 ; 2629-6055 (2019)
BASE
Show details
13
Assessing Composition in Sentence Vector Representations ...
BASE
Show details
14
Relating lexical and syntactic processes in language: Bridging research in humans and machines ...
Ettinger, Allyson. - : Digital Repository at the University of Maryland, 2018
BASE
Show details
15
Relating lexical and syntactic processes in language: Bridging research in humans and machines
BASE
Show details
16
Towards Linguistically Generalizable NLP Systems: A Workshop and Shared Task ...
BASE
Show details
17
The role of morphology in phoneme prediction: Evidence from MEG
In: Brain & language. - Orlando, Fla. [u.a.] : Elsevier 129 (2014), 14-23
OLC Linguistik
Show details
18
Mandarin utterance-final particle ba in the conversational scoreboard
In: LSA Annual Meeting Extended Abstracts; Vol 4: LSA Annual Meeting Extended Abstracts 2013; 13:1-5 ; 2377-3367 (2013)
BASE
Show details

Catalogues
0
0
1
0
0
0
0
Bibliographies
0
0
0
0
0
0
0
0
0
Linked Open Data catalogues
0
Online resources
0
0
0
0
Open access documents
17
0
0
0
0
© 2013 - 2024 Lin|gu|is|tik | Imprint | Privacy Policy | Datenschutzeinstellungen ändern