Towards Making a Dependency Parser See
Use este enlace para citar
http://hdl.handle.net/2183/36668Colecciones
- OpenAIRE [359]
- GI-LYS - Congresos, conferencias, etc. [71]
Metadatos
Mostrar el registro completo del ítemTítulo
Towards Making a Dependency Parser SeeFecha
2019-11Cita bibliográfica
Michalina Strzyz, David Vilares, and Carlos Gómez-Rodríguez. 2019. Towards Making a Dependency Parser See. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 1500–1506, Hong Kong, China. Association for Computational Linguistics.
Resumen
[Absctract]: We explore whether it is possible to leverage eye-tracking data in an RNN dependency parser (for English) when such information is only available during training - i.e. no aggregated or token-level gaze features are used at inference time. To do so, we train a multitask learning model that parses sentences as sequence labeling and leverages gaze features as auxiliary tasks. Our method also learns to train from disjoint datasets, i.e. it can be used to test whether already collected gaze features are useful to improve the performance on new non-gazed annotated treebanks. Accuracy gains are modest but positive, showing the feasibility of the approach. It can serve as a first step towards architectures that can better leverage eye-tracking data or other complementary information available only for training sentences, possibly leading to improvements in syntactic parsing.
Palabras clave
Eye-tracking data
RNN dependency parser
Multitask learning
Syntactic parsing
RNN dependency parser
Multitask learning
Syntactic parsing
Descripción
It was held in Hong Kong, China. 3-7 November, 2019
Versión del editor
Derechos
Atribución 3.0 España