Story Generation with Crowdsourced Plot Graphs

Li, Boyang; Lee-Urban, Stephen; Johnston, G. W.; Riedl, Mark O.

doi:10.1609/aaai.v27i1.8649

Cited by 122 publications

(58 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To cope with the content generation challenge in agent-based modeling of human social interaction [35], researchers explored different crowdsourcing techniques to gather social knowledge. Using, mainly, Amazon Mechanical Turk as a recruitment platform, workers are tasked with creating social exemplar datasets such as alibi stories [36] or small narratives [37]. [35] tries to diversify the generated dataset by integrating an improvisational theatre training technique with its crowdsourcing task.…”

Section: Data-driven Approach For Modelling Social Behaviourmentioning

confidence: 99%

Towards Explainable Social Agent Authoring tools: A case study on FAtiMA-Toolkit

Guimarães¹,

Campos²,

Santos³

et al. 2022

Preprint

View full text Add to dashboard Cite

Disclaimer: At the moment this article is currently awaiting submission review. Since this process usually takes a lot of time, feel free to use it this version as a reference until it is published. Thank you for reading, if you have any feedback please send us a message. Enjoy. The deployment of Socially Intelligent Agents (SIAs) in learning environments has proven to have several advantages in different areas of application. Social Agent Authoring Tools allow scenario designers to create tailored experiences with high control over SIAs behaviour, however, on the flip side, this comes at a cost as the complexity of the scenarios and its authoring can become overbearing. In this paper we introduce the concept of Explainable Social Agent Authoring Tools with the goal of analysing if authoring tools for social agents are understandable and interpretable. To this end we examine whether an authoring tool, FAtiMA-Toolkit, is understandable and its authoring steps interpretable, from the point-of-view of the author. We conducted two user studies to quantitatively assess the Interpretability, Comprehensibility and Transparency of FAtiMA-toolkit from the perspective of a scenario designer. One of the key findings is the fact that FAtiMA-Toolkit's conceptual model is, in general, understandable, however the emotional-based concepts were not as easily understood and used by the authors. Although there are some positive aspects regarding the explainability of FAtiMA-Toolkit, there is still progress to be made to achieve a fully explainable social agent authoring tool. We provide a set of key concepts and possible solutions that can guide developers to build such tools.SectionionIntroduction Socially Intelligent Agents (SIAs) have an ever increasing range of applications from conversational interfaces on websites to tutors or teammates in educational environments [1,2], where they are equipped with tools to conduct human-like interactions. Amongst the most promising applications of SIAs are serious games and social skills training environments. In these virtual environments SIAs behaviours can range from reactive wandering in the background of a scenario to complex social interactions that provide social support or assist the player in some skill training [3]. These autonomous agents sense the environment and act intelligently and independently from the user, allowing them to train and adapt specific verbal and nonverbal behaviors in socially challenging situations [4].

show abstract

Section: Data-driven Approach For Modelling Social Behaviourmentioning

confidence: 99%

Towards Explainable Social Agent Authoring tools: A case study on FAtiMA-Toolkit

Guimarães¹,

Campos²,

Santos³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…(Riedl and Harrison 2016) claims that stories are necessarily reflections of the culture and society; consequently, stories are a wealth of data where cultural values tacitly hold. They first generate a plot graph from crowdsourced stories using the technique described by (Li et al 2013). However, stories may not be detailed enough to describe sophisticated behavior such as driving cars.…”

Section: Learning-based Approachesmentioning

confidence: 99%

A Low-Cost Ethics Shaping Approach for Designing Reinforcement Learning Agents

Lin

2018

AAAI

View full text Add to dashboard Cite

This paper proposes a low-cost, easily realizable strategy to equip a reinforcement learning (RL) agent the capability of behaving ethically. Our model allows the designers of RL agents to solely focus on the task to achieve, without having to worry about the implementation of multiple trivial ethical patterns to follow. Based on the assumption that the majority of human behavior, regardless which goals they are achieving, is ethical, our design integrates human policy with the RL policy to achieve the target objective with less chance of violating the ethical code that human beings normally obey.

show abstract

“…On the other hand, LOT does not involve those tasks that require learning more particular features of stories, such as event chains (Chambers and Jurafsky, 2008), character types (Bamman et al, 2013), inter-character relations (Chaturvedi et al, 2016(Chaturvedi et al, , 2017, social networks (Agarwal et al, 2013), and abstractive structures (Finlayson, 2012). Non-neural story generation models usually retrieved events from a knowledge base with pre-specified semantic relations based on handcrafted rules (Li et al, 2013), which are costly and lack generalization. In this paper, we focus mainly on evaluating neural models for story understanding and generation.…”

Section: Introductionmentioning

confidence: 99%

LOT: A Story-Centric Benchmark for Evaluating Chinese Long Text Understanding and Generation

Guan¹,

Feng²,

Chen³

et al. 2022

Transactions of the Association for Computational Linguistics

View full text Add to dashboard Cite

Standard multi-task benchmarks are essential for developing pretraining models that can generalize to various downstream tasks. Existing benchmarks for natural language processing (NLP) usually focus only on understanding or generating short texts. However, long text modeling requires many distinct abilities in contrast to short texts, such as the modeling of long-range discourse and commonsense relations, and the coherence and controllability of generation. The lack of standardized benchmarks makes it difficult to assess these abilities of a model and fairly compare different models, especially Chinese models. Therefore, we propose a story-centric benchmark named LOT for evaluating Chinese long text modeling, which aggregates two understanding tasks and two generation tasks. We construct new datasets for these tasks based on human-written Chinese stories with hundreds of words. Furthermore, we release an encoder-decoder-based Chinese long text pretraining model named LongLM with up to 1 billion parameters. We pretrain LongLM on 120G Chinese novels with two generative tasks including text infilling and conditional continuation. Extensive experiments show that LongLM outperforms similar-sized pretraining models substantially on both the understanding and generation tasks in LOT.

show abstract

Story Generation with Crowdsourced Plot Graphs

Cited by 122 publications

References 9 publications

Towards Explainable Social Agent Authoring tools: A case study on FAtiMA-Toolkit

Towards Explainable Social Agent Authoring tools: A case study on FAtiMA-Toolkit

A Low-Cost Ethics Shaping Approach for Designing Reinforcement Learning Agents

LOT: A Story-Centric Benchmark for Evaluating Chinese Long Text Understanding and Generation

Contact Info

Product

Resources

About