Mexico

This paper concerns the evaluation of a workspace architecture for generating natural language descriptions, including methods for evaluating both its output and its own self-evaluation. Herein are details of preliminary results from evaluation of an early iteration of the architecture operating in the domain of weather. The domain is not typically seen as creative, but provides a simple testbed for the architecture and evaluation methodology. The program does not yet match humans in terms of fluency of language, factual correctness, and how completely the input is described, but human judges did find the program&#39;s output easier to read than human generated texts. Planned improvements to the program also described in the paper will incorporate self-monitoring and better self-evaluation with the aim of producing descriptions that are more fluently written and more accurate.

ICCC&#39;21

Evaluating Natural Language Descriptions Generated in a Workspace-Based Architecture

codelet

workspace

self-evaluation

self-monitoring

natural language generation

evaluation

The AACL, Asia-Pacific Chapter of the Association for Computational Linguistics, and the AFNLP, Asian Federation of Natural Language Processing, invite you to participate in the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (AACL) and the 10th International Joint Conference on Natural Language Processing (IJCNLP). 
<br>
<br>

A registration is required to access this event. Access to the event site will be granted to all registered attendees on **November 16th**. If you have not already done so, please register using the link below:
<br>
<br>https://conference.iis.sinica.edu.tw/servlet/Register?ConferenceID=411

Registration Is Required

IJCNLP-AACL 2022

The AACL, Asia-Pacific Chapter of the Association for Computational Linguistics, and the AFNLP, Asian Federation of Natural Language Processing, invite you to participate in the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (AACL) and the 10th International Joint Conference on Natural Language Processing (IJCNLP). 

This paper concerns the evaluation of a workspace architecture for generating natural language descriptions, including methods for evaluating both its output and its own self-evaluation. Herein are details of preliminary results from evaluation of an early iteration of the architecture operating in the domain of weather. The domain is not typically seen as creative, but provides a simple testbed for the architecture and evaluation methodology. The program does not yet match humans in terms of fluency of language, factual correctness, and how completely the input is described, but human judges did find the program's output easier to read than human generated texts. Planned improvements to the program also described in the paper will incorporate self-monitoring and better self-evaluation with the aim of producing descriptions that are more fluently written and more accurate.

technical paper

### **Welcome to the Twelfth International Conference on Computational Creativity, ICCC’21!**

<iframe width=1280 height=720 style="border:0;" scrolling="no" src="https://screencast-o-matic.com/player/crQbitV66dw?width=1280&height=720&ff=1&title=0" allowfullscreen="true"></iframe>

Computational creativity is an interdisciplinary endeavor that is located at the intersection of the fields of artificial intelligence, cognitive psychology, philosophy, social sciences and the arts. The goal of computational creativity is to model, simulate or replicate creativity using a computer, to achieve one of several ends:

* to construct a program or computer capable of showing some level of creativity
* to better understand human creativity and to formulate an algorithmic perspective on creative behavior in humans
* to design programs that can enhance human creativity without necessarily being creative themselves

<br>The ICCC conference series organized by the [**Association for Computational Creativity**](https://computationalcreativity.net/) since 2010 is the only scientific conference that focuses on computational creativity alone and also covers all aspects of it.



To access the Twelfth International Conference on Computational Creativity, ICCC’21, you need to register for the conference.  Instructions on how to access the event will be provided directly to attendees and included on the Twelfth International Conference on Computational Creativity, ICCC’21 conference website. 

[Register now](https://computationalcreativity.net/iccc21/registration/)

Register now

ICCC'21

Welcome to the Twelfth International Conference on Computational Creativity, ICCC’21!

Within the field of Computational Creativity, evaluation is one of the most important and more difficult tasks. Sometimes evaluation is part of the creative systems themselves, becoming an internal evaluation. Being a module of a creative system, it is useful to evaluate how effective this internal evaluation is.
In this paper, we propose a procedure for the (meta-)evaluation of internal evaluation modules, that allows for incremental development of both the evaluation module and the creative system, which are considered fully independent of each other. The procedure works by statistically comparing the evaluation of the average output of the generation system with the best results from the same, to see if the evaluation procedure can statistically distinguish the two. We then show how to apply the procedure giving one example evaluating a module we designed to assess structural coherence in generated folk music.

Meta-Evaluating Quantitative Internal Evaluation: a Practical Approach for Developers

Software design patterns can be helpful in describing the architecture of a system. Our objective is to obtain a broad overview of the current state-of-the-art of software design patterns used in Computational Creative (CC) systems. We conducted a systematic mapping study using manual and snowballing search techniques. Only 7 primary studies are identified in the CC community that explicitly mention the use of design patterns. Within these primary studies, 14 design patterns are mentioned, 12 of which are user-interaction design patterns rather than software design patterns describing the architecture of the system. The small number of primary studies indicates a gap in CC literature regarding the use of software design patterns in CC systems and motivates the need for research to identify software design patterns specific to CC systems.

Software Design Patterns of Computational Creativity: A Systematic Mapping Study

This paper describes a system for the design or redesign of movie posters. The main reasoning strategy used by the system is case-based reasoning (Leake 1996), which requires two important sub-tasks to be performed: case retrieval and case adaptation. We have used the random forest algorithm to implement the case retrieval subtask, as it allows the system to formulate a generalization of the features of all the top matching cases that are deter-mined to be most relevant to the new (re)design desired. We have used heuristic rules to implement the case ad-aptation task which results in generating the suggestion for poster composition. These heuristic rules are based on the Gestalt theory of composition (Arnheim 1974) and the requirements specified by the user.

A Case-Based Approach to Creating Movie Poster Compositions

In this paper we argue that the computational creativity community would benefit by working together to create a methodology for a systematic assessment and comparison of our creative agents. To achieve this goal, we suggest collaborating with experts in real-world environments to develop what we refer to as Rich Creative Environment Experience (RICEE). We describe a work in progress, where two artists and one CC scientist collaborate, to illustrate a possible way to build a RICEE. Based on this experience, we suggest some initial steps to start the development of this methodology. Thus, the purpose of this paper is to motivate the CC community to collaborate in the development of a methodology for the design and implementation of RICEEs.

The stone of madness meets AI

We conjecture that creativity and the perception of creativity are, at least to some extent, shaped by embodiment. This makes embodiment highly relevant for computational creativity (CC) research, but existing research is scarce and the use of the concept highly ambiguous. We overcome this situation by means of a systematic review and a prescriptive analysis of publications at the International Conference on Computational Creativity. We adopt and extend an established typology of embodiment to resolve ambiguity through identifying and comparing different usages of the concept. We collect, contextualise and highlight opportunities and challenges in embracing embodiment in CC as a reference for research, and put forward important directions to further the embodied CC research programme.

Embodiment and Computational Creativity (Study Paper)

In this presentation we focus on two of the major concepts we present in the main paper, namely: how we can understand the Typicality and Novelty of an object with Models created by an observer, and sophistication which lets us analyse the creator of a creative object. We also talk about highly sophisticated objects which can be regarded as masterpieces of the creator. Finally, we review some related works in computational aesthetics and how they can be improved. 



Incorporating Algorithmic Information Theory into Fundamental Concepts of Computational Creativity

Generative deep learning systems offer powerful tools for artefact generation, given their ability to model distributions of data and generate high-fidelity results. In the context of computational creativity, however, a major shortcoming is that they are unable to explicitly diverge from the training data in creative ways and are limited to fitting the target data distribution. To address these limitations, there have been a growing number of approaches for optimising, hacking and rewriting these models in order to actively diverge from the training data. We present a taxonomy and comprehensive survey of the state of the art of active divergence techniques, highlighting the potential for computational creativity researchers to advance these methods and use deep generative models in truly creative systems.

Active Divergence with Generative Deep Learning - A Survey and Taxonomy

Getting machines to generate text perceived as creative is a long-pursued goal. A growing body of research directs this goal towards augmenting the creative writing abilities of human authors. In this paper, we pursue this objective by analyzing how observing examples of automatically generated text influences writing. In particular, we examine a task referred to as sentence infilling, which involves transforming a list of words into a complete sentence. We emphasize “storiability” as a desirable feature of the resulting sentences, where “storiable” sentences are those that suggest a story a reader would be curious to hear about. Both humans and an automated system (based on a neural language model) performed this sentence infilling task. In one setting, people wrote sentences on their own; in a different setting, people observed the sentences produced by the model while writing their own sentences. Readers then assigned storiability preferences to the resulting sentences in a subsequent evaluation. We find that human-authored sentences were judged as more storiable when authors observed the generated examples, and that storiability increased as authors derived more semantic content from the examples. This result gives evidence of an “inspiration through observation” paradigm for human-computer collaborative writing, through which human writing can be enhanced by text generation models without directly copying their output.

Inspiration through Observation: Demonstrating the Influence of Automatically Generated Text on Creative Writing

Large pre-trained transformer-based language models have revolutionized the field of natural language processing in recent years. While BERT-like models perform exceptionally well for analytical tasks such as classification and regression, their text generation capabilities are usually limited to predicting tokens within a given context. In this paper, we introduce Galmet, a model that generates text by using genetic algorithms with BERT-like language models for evolving text. We use Galmet with the RoBERTa language model to automatically evolve real headlines into more satirical headlines. This is achieved by adapting the masked language head to the headlines domain for the mutation operator and finetuning a regression head to distinguish headlines from satire for the fitness function. We evaluated our system by comparing generated satirical headlines against human-edited headlines and just the fine-tuned masked language head. We found that while humans generally outperform the model, generations by Galmet are also often preferred over human-edited headlines. However, we also found that only using the fine-tuned masked language model gives slightly preferred satire due to generating more readable sentences. Galmet is thus a first step towards a new way of creating text generators using masked language models by transforming text guided by scores from another language model.

Survival of the Wittiest: Evolving Satire with Language Models

Recent interest in story generators capable of combining more than on plot line into an elaborate story have been handicapped by the lack of either theoretical material or quantitative metrics to ascertain the quality of outputs of such attempts. The present short paper postulates a set of metrics designed to capture some of the insights elaborated during a formative evaluation of an existing attempt at plot weaving.

Evaluating Natural Language Descriptions Generated in a Workspace-Based Architecture

Downloads

Next from ICCC'21

Meta-Evaluating Quantitative Internal Evaluation: a Practical Approach for Developers

Similar lecture

Assessing Neural Referential Form Selectors on a Realistic Multilingual Dataset

Evaluating Natural Language Descriptions Generated in a Workspace-Based Architecture

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ICCC'21

Meta-Evaluating Quantitative Internal Evaluation: a Practical Approach for Developers

Similar lecture

Assessing Neural Referential Form Selectors on a Realistic Multilingual Dataset

Downloads