Guided Neural Language Generation for Automated Storytelling

Abstract

Neural network based approaches to automated story plot generation attempt to learn how to generate novel plots from a corpus of natural language plot summaries. Prior work has shown that a semantic abstraction of sentences called events improves neural plot generation and and allows one to decompose the problem into: (1) the generation of a sequence of events (event-to-event) and (2) the transformation of these events into natural language sentences (event-to-sentence). However, typical neural language generation approaches to event-to-sentence can ignore the event details and produce grammaticallycorrect but semantically-unrelated sentences. We present an ensemble-based model that generates natural language guided by events. Our method outperforms the baseline sequence-tosequence model. Additionally, we provide results for a full end-to-end automated story generation system, demonstrating how our model works with existing systems designed for the event-to-event problem.

Where to Hide a Stolen Elephant: Leaps in Creative Writing with Multimodal Machine Intelligence

“…Fig. 2 shows both interfaces, including an active multimodal response in (B) with images 1 and sounds. Fig.…”

“…Recent work in machine learning, intended to improve performance on these tasks and others (such as machine translation and text summarization), has given rise to formidable natural language understanding and generation models [16,81]. These are often demonstrated by application to automated or semi-automated narrative generation tasks [1,62,76], an essentially creative domain. Given these advances, some recent work has begun to investigate the possibility for such models to be applied to enhancing human creativity, in a machine-in-the-loop setting [18,22].…”

Teaching Game AI as an Undergraduate Course in Computational Media

“…However, it was not represented in our programming assignments, and there is little 1 industry adoption of machine learning in Game AI. Nevertheless, we feel that future iterations of this course should demonstrate a role for machine learning, particularly where it might seem unexpected, such as in narrative generation (Ammanabrolu et al 2019).…”

THEaiTRE: Artificial Intelligence to Write a Theatre Play

“…The first line of work focuses on providing a higherlevel semantic representation to the networks and conditioning the generation on it. and Ammanabrolu et al [2019;2020] use an event-based representation, where an event roughly represents a clause (predicate, subject, direct and indirect object). The model generates the story at the event level and subsequently realizes the individual events to surface sentences.…”