DE eng

Search in the Catalogues and Directories

Hits 1 – 7 of 7

1
Incorporating Residual and Normalization Layers into Analysis of Masked Language Models ...
BASE
Show details
2
Topicalization in Language Models: A Case Study on Japanese ...
Abstract: Human-like syntactic generalization abilities of neural language models (LMs) have received wide attention, but discourse-level knowledge in these models has been less explored. In this study, we analyze the LMs with respect to a particular aspect of discourse, topic-comment structure, a crucial part of text production. To separately analyze the knowledge of LMs on topic-comment structure, we chose the Japanese language, a topic-prominent language, for designing probing tasks. Experimental results suggest that LMs have different generalizations from humans in terms of the topic-comment structure. These results imply that LMs do not perform human-like discourse processing in text generation. ...
Keyword: Computational Linguistics; Condensed Matter Physics; Electromagnetism; FOS Physical sciences; Information and Knowledge Engineering; Neural Network; Semantics
URL: https://underline.io/lecture/30067-topicalization-in-language-models-a-case-study-on-japanese
https://dx.doi.org/10.48448/c9h8-ys70
BASE
Hide details
3
Lower Perplexity is Not Always Human-Like ...
BASE
Show details
4
Lower Perplexity is Not Always Human-Like ...
BASE
Show details
5
Language Models as an Alternative Evaluator of Word Order Hypotheses: A Case Study in Japanese ...
BASE
Show details
6
Attention is Not Only a Weight: Analyzing Transformers with Vector Norms ...
BASE
Show details
7
Modeling Event Salience in Narratives via Barthes' Cardinal Functions ...
BASE
Show details

Catalogues
0
0
0
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
7
0
0
0
0
© 2013 - 2024 Lin|gu|is|tik | Imprint | Privacy Policy | Datenschutzeinstellungen ändern