Modeling Human Mental States with an Entity-based Narrative Graph

Autor:	Lee, I-Ta, Pacheco, Maria Leonor, Goldwasser, Dan
Rok vydání:	2021
Předmět:	Computer Science - Computation and Language Computer Science - Artificial Intelligence Computer Science - Machine Learning
Zdroj:	Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Druh dokumentu:	Working Paper
DOI:	10.18653/v1/2021.naacl-main.391
Popis:	Understanding narrative text requires capturing characters' motivations, goals, and mental states. This paper proposes an Entity-based Narrative Graph (ENG) to model the internal-states of characters in a story. We explicitly model entities, their interactions and the context in which they appear, and learn rich representations for them. We experiment with different task-adaptive pre-training objectives, in-domain training, and symbolic inference to capture dependencies between different decisions in the output space. We evaluate our model on two narrative understanding tasks: predicting character mental states, and desire fulfillment, and conduct a qualitative analysis. Comment: Accepted at NAACL 2021
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2104.07079 Zobrazit plný text záznamu View this record from Arxiv