Stori is currently focused on promoting financial inclusion in Mexico by providing underserved middle- and emerging middle-class customers with access to a credit card

Stori

stori

Decentralised AI Could Trade Up Lost Bitcoins (and Other Stories)

Authors:
(1) Gaurav Kolhatkar, SCTR’s Pune Institute of Computer Technology, Pune, India (gauravk403@gmail.com);
(2) Akshit Madan, SCTR’s Pune Institute of Computer Technology, Pune, India (akmadan17@gmail.com);
(3) Nidhi Kowtal, SCTR’s Pune Institute of Computer Technology, Pune, India (kowtalnidhi@gmail.com);
(4) Satyajit Roy, SCTR’s Pune Institute of Computer Technology, Pune, India (satyajit12.roy@gmail.com). Table of Links Abstract and Introduction Literature Survey Methodology Performance Analysis Conclusion and References Abstract—The conversion of user epics or stories into their appropriate representation in pseudocode or code is a time-consuming task, which can take up a large portion of the time in an industrial project. With this research paper, we aim to present a methodology to generate pseudocode from a given agile user story of small functionalities so as to reduce the overall time spent on the industrial project. Pseudocode is a programming language agnostic representation of the steps involved in a computer program, which can be easily converted into any programming language. Leveraging the potential of Natural Language Processing, we want to simplify the development process in organizations that use the Agile Model of Software Development. We present a methodology to convert a problem described in the English language into pseudocode. This methodology divides the Text to Pseudocode conversion task into two stages or subtasks, each of which is treated like an individual machine translation task. Stage 1 is Text to Code Conversion and Stage 2 is Code to Pseudocode Conversion. We find that the CodeT5 model gives the best results in terms of BLEU score when trained separately on the two subtasks mentioned above. BLEU score is a metric that is used to measure the similarity between a machine-translated text and a set of reference translations. Index Terms—Text to code generation, Code to Pseudocode generation, Transformers INTRODUCTION Efficiency of work is of the highest importance in modern organizations and businesses. A majority of the workplaces today use the Agile Model for software development. Agile is a software development approach based on iterative development, wherein tasks are divided into smaller iterations or sprints. In Agile project management tools such as Jira are used to document the user requirements in the form of epics or user stories. Developers need to understand these requirements and write code for the same. However, a significant amount of development time and efforts can be saved by automating the process of code/pseudo code generation, especially for simple or repetitive problems that have been solved before. The motivation of our research paper is to simplify the work of developers so that they can focus on more complex tasks and in the process, to optimize the software development lifecycle. Jira is a software application used for issue tracking and project management. It is widely used by agile development teams to track bugs, stories, epics, and other tasks. Epics are large bodies of work that can be broken down into a number of smaller tasks (called stories). Stories, also called “user stories,” are short requirements or requests written from the perspective of an end user. Our aim is to convert epics/stories to pseudo code. Despite the advantages of the Agile Model, developing software may still be a difficult and drawn-out process, especially when it comes to translating user requirements into functional code. Developers often translate user epics or stories manually into code as part of this process, which can take a lot of time and work. Our study intends to investigate the potential of utilising machine learning methods and natural language processing to automate the process of generating code and pseudocode from user stories in order to address this difficulty. By doing this, we hope to streamline developers’ tasks, enhance the software development lifecycle, and boost the effectiveness of the entire industrial project. Recent advancements in the field of natural language processing have made it possible to automate a variety of formerly manual operations. Recent developments in deep learning, in particular, have made it possible to create sophisticated natural language models that can extract context and meaning from text input. By utilising these models, we can quickly and efficiently create code or pseudocode from user stories, relieving the pressure on developers. Our study will examine a variety of currently used methodologies, including as rule-based systems, statistical models, and deep learning techniques, for producing code or pseudocode from user stories. We will assess the benefits and drawbacks of each methodology and suggest a novel strategy that makes the most of their advantages. Overall, our research paper aims to make software development more efficient and effective by automating the process of code/pseudocode generation from user stories. By doing so, we hope to free up developers to focus on more complex tasks, reduce the risk of errors, and ultimately deliver better software products to users. This paper is available on arxiv under CC 4.0 license. Authors: (1) Gaurav Kolhatkar, SCTR’s Pune Institute of Computer Technology, Pune, India (gauravk403@gmail.com); (2) Akshit Madan, SCTR’s Pune Institute of Computer Technology, Pune, India (akmadan17@gmail.com); (3) Nidhi Kowtal, SCTR’s Pune Institute of Computer Technology, Pune, India (kowtalnidhi@gmail.com); (4) Satyajit Roy, SCTR’s Pune Institute of Computer Technology, Pune, India (satyajit12.roy@gmail.com). Authors: Authors: (1) Gaurav Kolhatkar, SCTR’s Pune Institute of Computer Technology, Pune, India (gauravk403@gmail.com); SCTR’s Pune Institute of Computer Technology, (2) Akshit Madan, SCTR’s Pune Institute of Computer Technology, Pune, India (akmadan17@gmail.com); SCTR’s Pune Institute of Computer Technology, (3) Nidhi Kowtal, SCTR’s Pune Institute of Computer Technology, Pune, India (kowtalnidhi@gmail.com); SCTR’s Pune Institute of Computer Technology, (4) Satyajit Roy, SCTR’s Pune Institute of Computer Technology, Pune, India (satyajit12.roy@gmail.com). SCTR’s Pune Institute of Computer Technology, Table of Links Abstract and Introduction Abstract and Introduction Literature Survey Literature Survey Methodology Methodology Performance Analysis Performance Analysis Conclusion and References Conclusion and References Abstract—The conversion of user epics or stories into their appropriate representation in pseudocode or code is a time-consuming task, which can take up a large portion of the time in an industrial project. With this research paper, we aim to present a methodology to generate pseudocode from a given agile user story of small functionalities so as to reduce the overall time spent on the industrial project. Pseudocode is a programming language agnostic representation of the steps involved in a computer program, which can be easily converted into any programming language. Leveraging the potential of Natural Language Processing, we want to simplify the development process in organizations that use the Agile Model of Software Development. We present a methodology to convert a problem described in the English language into pseudocode. This methodology divides the Text to Pseudocode conversion task into two stages or subtasks, each of which is treated like an individual machine translation task. Stage 1 is Text to Code Conversion and Stage 2 is Code to Pseudocode Conversion. We find that the CodeT5 model gives the best results in terms of BLEU score when trained separately on the two subtasks mentioned above. BLEU score is a metric that is used to measure the similarity between a machine-translated text and a set of reference translations. Abstract —The conversion of user epics or stories into their appropriate representation in pseudocode or code is a time-consuming task, which can take up a large portion of the time in an industrial project. With this research paper, we aim to present a methodology to generate pseudocode from a given agile user story of small functionalities so as to reduce the overall time spent on the industrial project. Pseudocode is a programming language agnostic representation of the steps involved in a computer program, which can be easily converted into any programming language. Leveraging the potential of Natural Language Processing, we want to simplify the development process in organizations that use the Agile Model of Software Development. We present a methodology to convert a problem described in the English language into pseudocode. This methodology divides the Text to Pseudocode conversion task into two stages or subtasks, each of which is treated like an individual machine translation task. Stage 1 is Text to Code Conversion and Stage 2 is Code to Pseudocode Conversion. We find that the CodeT5 model gives the best results in terms of BLEU score when trained separately on the two subtasks mentioned above. BLEU score is a metric that is used to measure the similarity between a machine-translated text and a set of reference translations. Abstract Index Terms—Text to code generation, Code to Pseudocode generation, Transformers Index Terms —Text to code generation, Code to Pseudocode generation, Transformers Index Terms INTRODUCTION Efficiency of work is of the highest importance in modern organizations and businesses. A majority of the workplaces today use the Agile Model for software development. Agile is a software development approach based on iterative development, wherein tasks are divided into smaller iterations or sprints. In Agile project management tools such as Jira are used to document the user requirements in the form of epics or user stories. Developers need to understand these requirements and write code for the same. However, a significant amount of development time and efforts can be saved by automating the process of code/pseudo code generation, especially for simple or repetitive problems that have been solved before. The motivation of our research paper is to simplify the work of developers so that they can focus on more complex tasks and in the process, to optimize the software development lifecycle. Jira is a software application used for issue tracking and project management. It is widely used by agile development teams to track bugs, stories, epics, and other tasks. Epics are large bodies of work that can be broken down into a number of smaller tasks (called stories). Stories, also called “user stories,” are short requirements or requests written from the perspective of an end user. Our aim is to convert epics/stories to pseudo code. Despite the advantages of the Agile Model, developing software may still be a difficult and drawn-out process, especially when it comes to translating user requirements into functional code. Developers often translate user epics or stories manually into code as part of this process, which can take a lot of time and work. Our study intends to investigate the potential of utilising machine learning methods and natural language processing to automate the process of generating code and pseudocode from user stories in order to address this difficulty. By doing this, we hope to streamline developers’ tasks, enhance the software development lifecycle, and boost the effectiveness of the entire industrial project. Recent advancements in the field of natural language processing have made it possible to automate a variety of formerly manual operations. Recent developments in deep learning, in particular, have made it possible to create sophisticated natural language models that can extract context and meaning from text input. By utilising these models, we can quickly and efficiently create code or pseudocode from user stories, relieving the pressure on developers. Our study will examine a variety of currently used methodologies, including as rule-based systems, statistical models, and deep learning techniques, for producing code or pseudocode from user stories. We will assess the benefits and drawbacks of each methodology and suggest a novel strategy that makes the most of their advantages. Overall, our research paper aims to make software development more efficient and effective by automating the process of code/pseudocode generation from user stories. By doing so, we hope to free up developers to focus on more complex tasks, reduce the risk of errors, and ultimately deliver better software products to users. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Converting Epics/Stories into Pseudocode using Transformers

ট্রান্সফরমার ব্যবহার করে মহাকাব্য/গল্পকে সিউডোকোডে রূপান্তর করা

Konvertieren von Epics/Stories in Pseudocode mithilfe von Transformern

Conversión de epopeyas/historias en pseudocódigo usando Transformers

Conversion d'épopées/histoires en pseudocode à l'aide de Transformers

ट्रांसफॉर्मर्स का उपयोग करके महाकाव्यों/कहानियों को स्यूडोकोड में परिवर्तित करना

Transformers を使用してエピック/ストーリーを擬似コードに変換する

Transformers를 사용하여 에픽/스토리를 의사 코드로 변환

Convertendo épicos/histórias em pseudocódigo usando transformadores

Преобразование эпосов/историй в псевдокод с помощью трансформеров

Transformers Kullanarak Destanları/Hikayeleri Sahte Kod'a Dönüştürme

Chuyển đổi sử thi/câu chuyện thành mã giả bằng Transformers

使用 Transformers 将 Epics/Stories 转换为伪代码

Hey Hackers!Have you ever heard or seen the popular quote “There’s nothing new under the sun”?  How about this one “Ideas are unfaithful partners”? Both quotes point to the fact that, for the most part, everything we think, feel, or do has been thought, felt, or done by someone else at some point in the eternal crawl of time. This also applies to blog posts. Whether you’re writing a guide on how to create effective integration tests or an opinion piece on the AI bias problem, chances are that someone, somewhere has discussed a similar idea. This brings us to an important question: How do you float a unique POV in a sea of saturated subject matter? 1. Research, Research, and Some More Research for Good Measure This goes without saying, but just in case you need a reminder, research is arguably the most important part of a solid article. It’s an exercise that takes your ideas from mere speculation to valid theories. Before putting pen to paper (or fingers to keyboard), you must first research the topic at hand extensively. Find out what existing articles, on credible platforms, have to say about the topic. Look out for gaps, recent developments or emerging trends that haven’t been extensively covered yet. Fill those gaps! 2. Niche Down! Instead of trying to take on an entirely saturated subject matter like “What is AI?”, focus, instead, on a specific aspect or subtopic within it, like “How are Underdeveloped nations adjusting to the AI revolution?” This can help you find a unique perspective that hasn’t been addressed yet and position you as a thought leader in the field. Not to mention, it’s great for SEO purposes! 3. Lean on Stats and Data Nothing helps you stand out better than supporting your ideas with cold, hard data and statistics from credible sources. What’s more? The same data can tell different stories. For example, in a sample size of 10 people, if 50% like pineapples on pizza, you could frame the story in two ways. Story one: "5 out of 10 People Enjoy Pineapples on Pizza: Here’s Why." Story two: "Why 5 out of 10 People Dislike Pineapples on Pizza." Same data, two different narratives. 4. Analyze Contradictions or Controversies Sometimes, exploring contradictory viewpoints or controversial aspects of a topic can lead to interesting insights and a new angle. Here’s an example of two contradictory opinion pieces on the elusive identity of Satoshi Nakamoto: Craig Wright is Satoshi Nakamoto
Craig Wright is Not Satoshi Nakamoto 5. Think Outside the Box I know, I know. This is cliché advice. But, it is also important advice. When working on an idea that has been discussed extensively, always challenge yourself to look at it with a different lens. For example, if the topic is usually approached from a technical standpoint, consider exploring its social or ethical implications. And vice versa. Ready to breathe new life into an “old” story? Tell us what people aren’t saying about AI. The #bitcoin Writing Contest The #bitcoin Writing Contest, presented by Rootstock and HackerNoon, is now live with a total prize pool of $17,500 up for grabs!This once, you don’t have to worry about a new angle. Start writing today!! We’ve got some ideas laid out for you in this writing template. Summarily, for this contest, we welcome submissions on various Bitcoin topics like Runes, ordinals, layer 2 solutions, sidechains, and more. We're also interested in technical insights on SDKs, APIs, smart contract development frameworks, and tutorials for building applications on Rootstock. And that’s not all! There are more contests where that came from like The Crypto API Writing Contest by CoinGecko. You can find more details at contests.hackernoon.com Discover All HackerNoon Features on One Page We're constantly adding new features to help our writers create and share content effortlessly. Now, you can explore all these functionalities in one place! Visit our new Functionalities page! https://hackernoon.com/p/publish/functionality?embedable=true That’s all for today, Hackers.Until next time, keep reinventing the wheel! Hey Hackers! Have you ever heard or seen the popular quote “There’s nothing new under the sun”?  How about this one “Ideas are unfaithful partners”? Both quotes point to the fact that, for the most part, everything we think, feel, or do has been thought, felt, or done by someone else at some point in the eternal crawl of time. This also applies to blog posts. Whether you’re writing a guide on how to create effective integration tests or an opinion piece on the AI bias problem , chances are that someone, somewhere has discussed a similar idea. This brings us to an important question: create effective integration tests create effective integration tests AI bias problem AI bias problem How do you float a unique POV in a sea of saturated subject matter? How do you float a unique POV in a sea of saturated subject matter? How do you float a unique POV in a sea of saturated subject matter? 1. Research, Research, and Some More Research for Good Measure This goes without saying, but just in case you need a reminder, research is arguably the most important part of a solid article. It’s an exercise that takes your ideas from mere speculation to valid theories. Before putting pen to paper (or fingers to keyboard), you must first research the topic at hand extensively. Find out what existing articles, on credible platforms, have to say about the topic. Look out for gaps, recent developments or emerging trends that haven’t been extensively covered yet. Fill those gaps! 2. Niche Down! Instead of trying to take on an entirely saturated subject matter like “What is AI?”, focus, instead, on a specific aspect or subtopic within it, like “How are Underdeveloped nations adjusting to the AI revolution?” This can help you find a unique perspective that hasn’t been addressed yet and position you as a thought leader in the field. Not to mention, it’s great for SEO purposes! 3. Lean on Stats and Data Nothing helps you stand out better than supporting your ideas with cold, hard data and statistics from credible sources. What’s more? The same data can tell different stories. For example, in a sample size of 10 people, if 50% like pineapples on pizza, you could frame the story in two ways. Story one: "5 out of 10 People Enjoy Pineapples on Pizza: Here’s Why." Story two: "Why 5 out of 10 People Dislike Pineapples on Pizza." Same data, two different narratives. 4. Analyze Contradictions or Controversies 4. Analyze Contradictions or Controversies Sometimes, exploring contradictory viewpoints or controversial aspects of a topic can lead to interesting insights and a new angle. Here’s an example of two contradictory opinion pieces on the elusive identity of Satoshi Nakamoto: Craig Wright is Satoshi Nakamoto Craig Wright is Not Satoshi Nakamoto Craig Wright is Satoshi Nakamoto Craig Wright is Satoshi Nakamoto Craig Wright is Satoshi Nakamoto Craig Wright is Not Satoshi Nakamoto Craig Wright is Not Satoshi Nakamoto Craig Wright is Not Satoshi Nakamoto 5. Think Outside the Box 5. Think Outside the Box I know, I know. This is cliché advice. But, it is also important advice. When working on an idea that has been discussed extensively, always challenge yourself to look at it with a different lens. For example, if the topic is usually approached from a technical standpoint, consider exploring its social or ethical implications. And vice versa. Ready to breathe new life into an “old” story? Tell us what people aren’t saying about AI. Ready to breathe new life into an “old” story? Tell us what people aren’t saying about AI. Tell us what people aren’t saying about AI. Tell us what people aren’t saying about AI. The #bitcoin Writing Contest The #bitcoin Writing Contest The #bitcoin Writing Contest The #bitcoin Writing Contest , presented by Rootstock and HackerNoon, is now live with a total prize pool of $17,500 up for grabs! This once, you don’t have to worry about a new angle. #bitcoin Writing Contest #bitcoin Writing Contest Start writing today!! We’ve got some ideas laid out for you in this writing template. Start writing today!! We’ve got some ideas laid out for you in this writing template . writing template writing template Summarily, for this contest, we welcome submissions on various Bitcoin topics like Runes, ordinals, layer 2 solutions, sidechains, and more. We're also interested in technical insights on SDKs, APIs, smart contract development frameworks, and tutorials for building applications on Rootstock. And that’s not all! There are more contests where that came from like The Crypto API Writing Contest by CoinGecko . The Crypto API Writing Contest by CoinGecko The Crypto API Writing Contest by CoinGecko You can find more details at contests.hackernoon.com contests.hackernoon.com contests.hackernoon.com Discover All HackerNoon Features on One Page Discover All HackerNoon Features on One Page We're constantly adding new features to help our writers create and share content effortlessly. Now, you can explore all these functionalities in one place! Visit our new Functionalities page ! Functionalities page Functionalities page https://hackernoon.com/p/publish/functionality?embedable=true https://hackernoon.com/p/publish/functionality?embedable=true That’s all for today, Hackers. Until next time, keep reinventing the wheel!

How to Make Old Stories New Again: Insights from HackerNoon Editors

কীভাবে পুরানো গল্পগুলি আবার নতুন করা যায়: হ্যাকারনুন সম্পাদকদের কাছ থেকে অন্তর্দৃষ্টি

Wie man alten Geschichten wieder neues Leben einhaucht: Einblicke der HackerNoon-Redakteure

Cómo hacer que las viejas historias vuelvan a ser nuevas: ideas de los editores de HackerNoon

Comment recréer de vieilles histoires : aperçus des éditeurs de HackerNoon

पुरानी कहानियों को फिर से नया कैसे बनाएं: हैकरनून संपादकों की जानकारी

古いストーリーを再び新しくする方法: HackerNoon 編集者からの洞察

오래된 이야기를 다시 새로운 것으로 만드는 방법: HackerNoon 편집자의 통찰력

Como transformar histórias antigas em novas novamente: insights dos editores do HackerNoon

Как сделать старые истории снова новыми: советы редакторов HackerNoon

Eski Hikayeler Nasıl Yeniden Yeni Hale Getirilir: HackerNoon Editörlerinden İçgörüler

Cách làm mới những câu chuyện cũ: Thông tin chi tiết từ các biên tập viên HackerNoon

如何让旧故事焕发新生：HackerNoon 编辑的见解

Hey Hackers! HackerNoon wouldn’t be HackerNoon if we didn’t spend time updating our CMS with thoughtful upgrades and new writer flows! By now, most of you must have already noticed a new look to the writer user interface so let’s dive into the changes. Tags Placement Has Changed The first change you’ll notice is that the tags placement box now lives under the headline. Tags are an important part of the HackerNoon writer experience and can help you 100x your story views. HackerNoon wants to emphasize the importance of tags with this change in placement. As a reminder, you should always prefer adding new and unique long tail keywords as opposed to generic ones to get the most out of your story and have it distributed far and wide on the internet. New Submission Flow The biggest change in the writer's UI is the submission process. HackerNoon now incentivizes authors who post original content with faster turnaround times. By default, the “submit story” button no longer appears unless you specify whether your story is an original submission i.e. not a repost or not (Remember: no cheating. Our Plagiarism checker is quite robust 😉) If your story is not a repost - meaning, it’s an original submission - simply hit “yes” and submit your story for review! So does that mean HackerNoon is not accepting reposts? Of course not! Reposts are still welcome on HackerNoon! When submitting a repost, select “no” and then enter the URL of the website where the story was originally published. Once you do, simply specify whether the URL is a personal website or a business blog. Websites like Medium, Dev.to, Hashnode, Substack are considered personal posts, and fall under the ambit of a personal website. Once you specify that the URL is personal, you can submit the story! But what happens if you represent a brand and want to repost from your brand blog? Simply specify that the URL is a business site and watch the magic happen 😉 Jokes aside, brands need to purchase credits on HackerNoon to publish and the updated writer flow makes it easier for businesses looking to post on HackerNoon to purchase credits to do so! And that’s that! That’s it for the major changes to the HackerNoon user interface. Hope these explanations help and looking forward to future submissions! Hey Hackers! HackerNoon wouldn’t be HackerNoon if we didn’t spend time updating our CMS with thoughtful upgrades and new writer flows! By now, most of you must have already noticed a new look to the writer user interface so let’s dive into the changes. Tags Placement Has Changed The first change you’ll notice is that the tags placement box now lives under the headline. Tags are an important part of the HackerNoon writer experience and can help you 100x your story views. HackerNoon wants to emphasize the importance of tags with this change in placement. As a reminder, you should always prefer adding new and unique long tail keywords as opposed to generic ones to get the most out of your story and have it distributed far and wide on the internet. New Submission Flow The biggest change in the writer's UI is the submission process. HackerNoon now incentivizes authors who post original content with faster turnaround times. By default, the “submit story” button no longer appears unless you specify whether your story is an original submission i.e. not a repost or not (Remember: no cheating. Our Plagiarism checker is quite robust 😉) If your story is not a repost - meaning, it’s an original submission - simply hit “yes” and submit your story for review! So does that mean HackerNoon is not accepting reposts? Of course not! Reposts are still welcome on HackerNoon! When submitting a repost, select “no” and then enter the URL of the website where the story was originally published. Once you do, simply specify whether the URL is a personal website or a business blog. Websites like Medium, Dev.to, Hashnode, Substack are considered personal posts, and fall under the ambit of a personal website. Once you specify that the URL is personal, you can submit the story! But what happens if you represent a brand and want to repost from your brand blog? Simply specify that the URL is a business site and watch the magic happen 😉 Jokes aside, brands need to purchase credits on HackerNoon to publish and the updated writer flow makes it easier for businesses looking to post on HackerNoon to purchase credits to do so! And that’s that! That’s it for the major changes to the HackerNoon user interface. Hope these explanations help and looking forward to future submissions!

Understand the New HackerNoon UI and Let’s Get Submittin’ (Some Cool Stories)

নতুন হ্যাকারনুন UI বুঝুন এবং আসুন সাবমিটিন পান' (কিছু দুর্দান্ত গল্প)

Lernen Sie die neue HackerNoon-Benutzeroberfläche kennen und legen Sie los (ein paar coole Geschichten)

Comprenda la nueva interfaz de usuario de HackerNoon y comencemos a enviar (algunas historias interesantes)

Comprenez la nouvelle interface utilisateur HackerNoon et passons à la soumission (quelques histoires sympas)

नए HackerNoon UI को समझें और सबमिट करना शुरू करें (कुछ अच्छी कहानियाँ)

新しい HackerNoon UI を理解して、提出してみましょう (クールなストーリー)

새로운 HackerNoon UI를 이해하고 제출해 봅시다(몇 가지 멋진 이야기)

Entenda a nova interface do HackerNoon e vamos enviar (algumas histórias interessantes)

Разберитесь с новым пользовательским интерфейсом HackerNoon и приступим к отправке (несколько интересных историй)

Yeni HackerNoon Kullanıcı Arayüzünü Anlayın ve Gönderime Başlayalım (Bazı Harika Hikayeler)

Tìm hiểu giao diện người dùng HackerNoon mới và hãy cùng gửi đi (Một số câu chuyện thú vị)

了解新的 HackerNoon UI 并开始提交（一些有趣的故事）

Authors:
(1) PIOTR MIROWSKI and KORY W. MATHEWSON, DeepMind, United Kingdom and Both authors contributed equally to this research;
(2) JAYLEN PITTMAN, Stanford University, USA and Work done while at DeepMind;
(3) RICHARD EVANS, DeepMind, United Kingdom. Table of Links Abstract and Intro Storytelling, The Shape of Stories, and Log Lines The Use of Large Language Models for Creative Text Generation Evaluating Text Generated by Large Language Models Participant Interviews Participant Surveys Discussion and Future Work Conclusions, Acknowledgements, and References A. RELATED WORK ON AUTOMATED STORY GENERATION AND CONTROLLABLE STORY GENERATION B. ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM C. DETAILS OF QUANTITATIVE OBSERVATIONS D. SUPPLEMENTARY FIGURES E. FULL PROMPT PREFIXES FOR DRAMATRON F. RAW OUTPUT GENERATED BY DRAMATRON G. CO-WRITTEN SCRIPTS B ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM Discussions amongst the creative team were summarized in the body of the text (see Section 5.9). To reiterate, four key themes emerged through these discussions which echo the themes presented in Section 5. These themes are discussed in detail in this section, alongside supporting quotes. First, the system has a distinct glitch style that can sometimes be nonsensical, vague, or passive. As one performer recounted, “sometimes there is internal conflict within the text, and in those moments it is as if [Dramatron] is working against itself”. This pushes performers to commit to choices, and to be concise and specific and complimentary in their interpretation. One performer noted that if the system generated flawless text, “it might not work as well”, because “the mistakes in the script are the joy”. Another went a step further, saying “I almost want more curveballs and non-sequiturs... more crunchy bits”. Overall, the sentiment of enjoying the style of the system was a common theme, with several of the performers remarking that “some of the funniest parts are when you can tell a robot made it”, and that the audience “wants to hear the robot’s voice”. Secondly, the generated text can sometimes be repetitive. This can be interpreted as a mistake. Or, this repetition can be fun and playful if it is interpreted as an important and deliberate choice from the playwright: “a choice to be honored,” as one performer said. When the system did repeat itself, the cast was able to make the lines more meaningful with their own unique human talents. As one said, “you can do so much with your line reading, with your physicality, delivery, proximity, and acting choices.” Third, the team discussed agency and expectations of the systems capabilities. For instance, in the way that the performers would refer to the system’s choices. One said “Dramatron was trying to show me the angle,” or “I was trying to understand what Dramatron meant”. Discussions amongst the creative team explored their expectations of what the system could and could not do. For example, “it is just a bot, not really fully understanding how [the world] works”, and another responded “Dramatron is trying.” Finally, the prevailing feedback from the majority of performers was that participating in the production was fun. As one actor said, “it was liberating and easy because the world creating was done for you, the platform was done, and that can be mentally exhausting as an improviser.” These reflections discussed by the creative team reflect the usefulness of co-written scripts. This is particularly true when used for a production such as Plays By Bots, which leverages professional improvisers to interpret the scripts. The co-creativity of a system such as Dramatron extends beyond the playwright and Dramatron, and to the performer on the stage working with the generated and edited text. These reflections discussed by the creative team reflect the usefulness of co-written scripts. This paper is available on arxiv under CC 4.0 license. Authors: (1) PIOTR MIROWSKI and KORY W. MATHEWSON, DeepMind, United Kingdom and Both authors contributed equally to this research; (2) JAYLEN PITTMAN, Stanford University, USA and Work done while at DeepMind; (3) RICHARD EVANS, DeepMind, United Kingdom. Authors: Authors: (1) PIOTR MIROWSKI and KORY W. MATHEWSON, DeepMind, United Kingdom and Both authors contributed equally to this research; (2) JAYLEN PITTMAN, Stanford University, USA and Work done while at DeepMind; (3) RICHARD EVANS, DeepMind, United Kingdom. Table of Links Abstract and Intro Abstract and Intro Storytelling, The Shape of Stories, and Log Lines Storytelling, The Shape of Stories, and Log Lines The Use of Large Language Models for Creative Text Generation The Use of Large Language Models for Creative Text Generation Evaluating Text Generated by Large Language Models Evaluating Text Generated by Large Language Models Participant Interviews Participant Interviews Participant Surveys Participant Surveys Discussion and Future Work Discussion and Future Work Conclusions, Acknowledgements, and References Conclusions, Acknowledgements, and References A. RELATED WORK ON AUTOMATED STORY GENERATION AND CONTROLLABLE STORY GENERATION A. RELATED WORK ON AUTOMATED STORY GENERATION AND CONTROLLABLE STORY GENERATION B. ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM B. ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM C. DETAILS OF QUANTITATIVE OBSERVATIONS C. DETAILS OF QUANTITATIVE OBSERVATIONS D. SUPPLEMENTARY FIGURES D. SUPPLEMENTARY FIGURES E. FULL PROMPT PREFIXES FOR DRAMATRON E. FULL PROMPT PREFIXES FOR DRAMATRON F. RAW OUTPUT GENERATED BY DRAMATRON F. RAW OUTPUT GENERATED BY DRAMATRON G. CO-WRITTEN SCRIPTS G. CO-WRITTEN SCRIPTS B ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM Discussions amongst the creative team were summarized in the body of the text (see Section 5.9). To reiterate, four key themes emerged through these discussions which echo the themes presented in Section 5. These themes are discussed in detail in this section, alongside supporting quotes. First, the system has a distinct glitch style that can sometimes be nonsensical, vague, or passive. As one performer recounted, “sometimes there is internal conflict within the text, and in those moments it is as if [Dramatron] is working against itself”. This pushes performers to commit to choices, and to be concise and specific and complimentary in their interpretation. One performer noted that if the system generated flawless text, “it might not work as well”, because “the mistakes in the script are the joy”. Another went a step further, saying “I almost want more curveballs and non-sequiturs... more crunchy bits”. Overall, the sentiment of enjoying the style of the system was a common theme, with several of the performers remarking that “some of the funniest parts are when you can tell a robot made it”, and that the audience “wants to hear the robot’s voice”. Secondly, the generated text can sometimes be repetitive. This can be interpreted as a mistake. Or, this repetition can be fun and playful if it is interpreted as an important and deliberate choice from the playwright: “a choice to be honored,” as one performer said. When the system did repeat itself, the cast was able to make the lines more meaningful with their own unique human talents. As one said, “you can do so much with your line reading, with your physicality, delivery, proximity, and acting choices.” Third, the team discussed agency and expectations of the systems capabilities. For instance, in the way that the performers would refer to the system’s choices. One said “Dramatron was trying to show me the angle,” or “I was trying to understand what Dramatron meant”. Discussions amongst the creative team explored their expectations of what the system could and could not do. For example, “it is just a bot, not really fully understanding how [the world] works”, and another responded “Dramatron is trying.” Finally, the prevailing feedback from the majority of performers was that participating in the production was fun. As one actor said, “it was liberating and easy because the world creating was done for you, the platform was done, and that can be mentally exhausting as an improviser.” These reflections discussed by the creative team reflect the usefulness of co-written scripts. This is particularly true when used for a production such as Plays By Bots, which leverages professional improvisers to interpret the scripts. The co-creativity of a system such as Dramatron extends beyond the playwright and Dramatron, and to the performer on the stage working with the generated and edited text. These reflections discussed by the creative team reflect the usefulness of co-written scripts. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

How Exactly Does Dramatron Cook Stories Up

Authors:
(1) PIOTR MIROWSKI and KORY W. MATHEWSON, DeepMind, United Kingdom and Both authors contributed equally to this research;
(2) JAYLEN PITTMAN, Stanford University, USA and Work done while at DeepMind;
(3) RICHARD EVANS, DeepMind, United Kingdom. Table of Links Abstract and Intro Storytelling, The Shape of Stories, and Log Lines The Use of Large Language Models for Creative Text Generation Evaluating Text Generated by Large Language Models Participant Interviews Participant Surveys Discussion and Future Work Conclusions, Acknowledgements, and References A. RELATED WORK ON AUTOMATED STORY GENERATION AND CONTROLLABLE STORY GENERATION B. ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM C. DETAILS OF QUANTITATIVE OBSERVATIONS D. SUPPLEMENTARY FIGURES E. FULL PROMPT PREFIXES FOR DRAMATRON F. RAW OUTPUT GENERATED BY DRAMATRON G. CO-WRITTEN SCRIPTS 2 STORYTELLING, THE SHAPE OF STORIES, AND LOG LINES Archaeological evidence including artists’ depictions of narratives on jars from as far back as 2400 BCE, show how ancient societies used the technologies of their time to help tell their stories [43]. In Poetics [5], Aristotle identified plot (or mythos) as the most important part of a tragic story—the other elements being characters, their thoughts, language, melody and spectacle. The plot is the sequence of actions that shape the story. Each plot point is coherent with, and a consequence of, the previous point(s). Aristotle introduced the well-known simple 3-point plot: beginning, middle, and end. There are many dramatic structures extending the work of Aristotle [82]. For example, one of the many adaptations of Freytag’s pyramid [40], popular in Western storytelling is illustrated in Fig. 2. It includes this sequence of plot points: Exposition, Inciting Incident, a Rising Action composed of a series of Conflicts, Complications, and Dilemmas, Climax, Falling Action, Resolution, and Dénouement. Seminal work in narrative analysis [58, 83, 93] suggests general but mutable structures present in storytelling processes across many linguistic and cultural traditions (e.g. abstract, orientation, complicating action, and coda narrative clauses). The finding that narratives in many societies and in various languages make use of large, modular elements has inspired the use of models of narrative in automated storytelling, as can be seen in fields such as computational narratology [53, 63]. The general structures found in the Structuralist narratology of Propp [83] and the Personal Experience Narratives of Labov and Waletzky [58] are aligned with Freytag’s pyramid, which we choose because it is more in line with the specific discourse genre of dramatic scripts, and arguably more familiar to the playwrights we engaged with over the course of our study. However, we note that our choice of narrative structure is not universal in scope, and is, in fact, "characteristically Western" [26, 53]. Alternative story shapes, or "story grammars" [94] are possible [6, 17]. Fig. 2 also shows an alternative story structure: the Hero’s Journey or Monomyth [15, 110]. For narrative coherence, we leverage plot beats in our hierarchical generation. Oftentimes the seed of narrative inspiration for a screenplay or theatre script is a log line [98]. The log line summarizes the story in a few sentences and is often adapted over the course of production due to creative team preferences and cultural references. Log lines typically contain the setting, protagonist, antagonist, a conflict or goal, and sometimes the inciting incident [8]. In this work, we use log lines to start the hierarchical story generation process because it contains plot elements in the answers to questions: Who? What? When and Where? How? Why? [101] This paper is available on arxiv under CC 4.0 license. Authors: (1) PIOTR MIROWSKI and KORY W. MATHEWSON, DeepMind, United Kingdom and Both authors contributed equally to this research; (2) JAYLEN PITTMAN, Stanford University, USA and Work done while at DeepMind; (3) RICHARD EVANS, DeepMind, United Kingdom. Authors: Authors: (1) PIOTR MIROWSKI and KORY W. MATHEWSON, DeepMind, United Kingdom and Both authors contributed equally to this research; (2) JAYLEN PITTMAN, Stanford University, USA and Work done while at DeepMind; (3) RICHARD EVANS, DeepMind, United Kingdom. Table of Links Abstract and Intro Abstract and Intro Storytelling, The Shape of Stories, and Log Lines Storytelling, The Shape of Stories, and Log Lines The Use of Large Language Models for Creative Text Generation The Use of Large Language Models for Creative Text Generation Evaluating Text Generated by Large Language Models Evaluating Text Generated by Large Language Models Participant Interviews Participant Interviews Participant Surveys Participant Surveys Discussion and Future Work Discussion and Future Work Conclusions, Acknowledgements, and References Conclusions, Acknowledgements, and References A. RELATED WORK ON AUTOMATED STORY GENERATION AND CONTROLLABLE STORY GENERATION A. RELATED WORK ON AUTOMATED STORY GENERATION AND CONTROLLABLE STORY GENERATION B. ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM B. ADDITIONAL DISCUSSION FROM PLAYS BY BOTS CREATIVE TEAM C. DETAILS OF QUANTITATIVE OBSERVATIONS C. DETAILS OF QUANTITATIVE OBSERVATIONS D. SUPPLEMENTARY FIGURES D. SUPPLEMENTARY FIGURES E. FULL PROMPT PREFIXES FOR DRAMATRON E. FULL PROMPT PREFIXES FOR DRAMATRON F. RAW OUTPUT GENERATED BY DRAMATRON F. RAW OUTPUT GENERATED BY DRAMATRON G. CO-WRITTEN SCRIPTS G. CO-WRITTEN SCRIPTS 2 STORYTELLING, THE SHAPE OF STORIES, AND LOG LINES Archaeological evidence including artists’ depictions of narratives on jars from as far back as 2400 BCE, show how ancient societies used the technologies of their time to help tell their stories [43]. In Poetics [5], Aristotle identified plot (or mythos) as the most important part of a tragic story—the other elements being characters, their thoughts, language, melody and spectacle. The plot is the sequence of actions that shape the story. Each plot point is coherent with, and a consequence of, the previous point(s). Aristotle introduced the well-known simple 3-point plot: beginning, middle, and end. There are many dramatic structures extending the work of Aristotle [82]. For example, one of the many adaptations of Freytag’s pyramid [40], popular in Western storytelling is illustrated in Fig. 2. It includes this sequence of plot points: Exposition, Inciting Incident, a Rising Action composed of a series of Conflicts, Complications, and Dilemmas, Climax, Falling Action, Resolution, and Dénouement. Seminal work in narrative analysis [58, 83, 93] suggests general but mutable structures present in storytelling processes across many linguistic and cultural traditions (e.g. abstract, orientation, complicating action, and coda narrative clauses). The finding that narratives in many societies and in various languages make use of large, modular elements has inspired the use of models of narrative in automated storytelling, as can be seen in fields such as computational narratology [53, 63]. The general structures found in the Structuralist narratology of Propp [83] and the Personal Experience Narratives of Labov and Waletzky [58] are aligned with Freytag’s pyramid, which we choose because it is more in line with the specific discourse genre of dramatic scripts, and arguably more familiar to the playwrights we engaged with over the course of our study. However, we note that our choice of narrative structure is not universal in scope, and is, in fact, "characteristically Western" [26, 53]. Alternative story shapes, or "story grammars" [94] are possible [6, 17]. Fig. 2 also shows an alternative story structure: the Hero’s Journey or Monomyth [15, 110]. For narrative coherence, we leverage plot beats in our hierarchical generation. Oftentimes the seed of narrative inspiration for a screenplay or theatre script is a log line [98]. The log line summarizes the story in a few sentences and is often adapted over the course of production due to creative team preferences and cultural references. Log lines typically contain the setting, protagonist, antagonist, a conflict or goal, and sometimes the inciting incident [8]. In this work, we use log lines to start the hierarchical story generation process because it contains plot elements in the answers to questions: Who? What? When and Where? How? Why? [101] This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Storytelling, The Shape of Stories, and Log Lines

Understanding and Generating Dialogue between Characters in Stories: Limitations and References

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 9 Conclusion In this work, we present the first study on understanding and generating inter-character dialogue in stories. To this end, we collect a Chinese story dataset DIALSTORY with a large amount of dialogue, and propose two new tasks including masked dialogue generation and dialogue speaker recognition. We also construct standardized datasets for these tasks through automatic and manual annotations based on DIALSTORY. By incorporating representations of different characters, our model outperforms strong baselines significantly on both tasks in terms of automatic and manual evaluation. The benchmark datasets, tasks, and models will further boost the development of this field. This paper is available on arxiv under CC 4.0 DEED license. Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 9 Conclusion In this work, we present the first study on understanding and generating inter-character dialogue in stories. To this end, we collect a Chinese story dataset DIALSTORY with a large amount of dialogue, and propose two new tasks including masked dialogue generation and dialogue speaker recognition. We also construct standardized datasets for these tasks through automatic and manual annotations based on DIALSTORY. By incorporating representations of different characters, our model outperforms strong baselines significantly on both tasks in terms of automatic and manual evaluation. The benchmark datasets, tasks, and models will further boost the development of this field. This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv

Understanding and Generating Dialogue between Characters in Stories: Conclusion

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 8 Future Work Despite the overall improved performance of our character modeling methods, we find that there is still space for further improvement. Our approach constructs the character representations from the input stories as static vectors. But the status of characters could be updated during the generation process. As a result, we can dynamically update those representations during plot development. We plan to explore the new possibility of the character update technique and leave it as an important future work. This paper is available on arxiv under CC 4.0 DEED license. Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 8 Future Work Despite the overall improved performance of our character modeling methods, we find that there is still space for further improvement. Our approach constructs the character representations from the input stories as static vectors. But the status of characters could be updated during the generation process. As a result, we can dynamically update those representations during plot development. We plan to explore the new possibility of the character update technique and leave it as an important future work. This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv

Understanding and Generating Dialogue between Characters in Stories: Future Work

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 7 Discussion Masked Dialogue Generation In this task, the masked positions can be anywhere in the story. To generate and complete one specific masked dialogue turn, this task requires the machine to first understand the main line of the whole story, the roles and features of different characters, and then infer and generate the most appropriate dialogue turn to move the plot forward according to the specific environment. Context information also plays an important role in this task, which puts forth a high demand for dialogue and plot coherence. And this task can easily leverage other research such as persona chat bots, emotional chat bots, or even some other story generation tasks. Dialogue Speaker Recognition This task specifically focuses on the dialogue understanding in a story. In a story, there are always lots of characters speaking to each other in different environments. And the relationship between characters and dialogue is complicated. Sometimes a piece of dialogue is self-talk, sometimes it is a conversation between two parties, or even multiple parties in a complex environment, and sometimes there are many details in the background information that cover up the actual speaker. The speaker of the dialogue will not only appear in the surrounding, but may also hide in the previous or behind the context of the dialogue. To recognize all the speakers in one story, our model also needs to capture different characteristics of people. Higher requirements for the model’s understanding of dialogue and character relationships are also put forth by the necessity to predict speakers simultaneously in several positions. This task can also leverage other research on recognition in multi-person dialogue story scenes. This paper is available on arxiv under CC 4.0 DEED license. Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 7 Discussion Masked Dialogue Generation In this task, the masked positions can be anywhere in the story. To generate and complete one specific masked dialogue turn, this task requires the machine to first understand the main line of the whole story, the roles and features of different characters, and then infer and generate the most appropriate dialogue turn to move the plot forward according to the specific environment. Context information also plays an important role in this task, which puts forth a high demand for dialogue and plot coherence. And this task can easily leverage other research such as persona chat bots, emotional chat bots, or even some other story generation tasks. Masked Dialogue Generation Dialogue Speaker Recognition This task specifically focuses on the dialogue understanding in a story. In a story, there are always lots of characters speaking to each other in different environments. And the relationship between characters and dialogue is complicated. Sometimes a piece of dialogue is self-talk, sometimes it is a conversation between two parties, or even multiple parties in a complex environment, and sometimes there are many details in the background information that cover up the actual speaker. The speaker of the dialogue will not only appear in the surrounding, but may also hide in the previous or behind the context of the dialogue. To recognize all the speakers in one story, our model also needs to capture different characteristics of people. Higher requirements for the model’s understanding of dialogue and character relationships are also put forth by the necessity to predict speakers simultaneously in several positions. This task can also leverage other research on recognition in multi-person dialogue story scenes. Dialogue Speaker Recognition This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv

Understanding and Generating Dialogue between Characters in Stories: Discussion

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 6 Experiments 6.1 Masked Dialogue Generation Implementation Details To conduct experiments on the masked dialogue generation task, we decide the hyper-parameters based on the performance of the validation set. We train Shao et al. (2021)’s BART model for 4.6 epochs with a 1e-4 learning rate for 1 day, and for our model, we train it for 5.6 epochs with a 1e-4 learning rate for 1 day. All baselines and our model are trained using the Adam optimizer. During the training process for our method, we computed the selection coverage of characters within a single story. And it showed that in every 1000 training steps, the coverage of different characters ranged from 98.64% to 99.00%, which meant nearly all the characters are selected during training, and all the characters contributed to the generated dialogue. It further proved that the argmax in Eq. 3 operation doesn’t break the gradient progress when training for this task. Automatic Evaluation Following previous works, we use several standard, widely used automatic evaluation metrics. We use BLEUn (Papineni et al., 2002) to measure the average word overlap between each generated and groundtruth dialogue turn (n=1,2), and Distinct-n (Li et al., 2015) to evaluate n-gram diversity of generated dialogue turns (n=2,3,4). To be more specific, for the coherence classifier, we construct the training and validation sets by randomly shuffling the order of dialogue turns and keeping other content in the correct order. We regard the perturbed story as a negative example and the original story as a positive example. We sample another 195k stories (except those in DIALSTORY) from the novels of Guan et al. (2022) to construct the training set (190k examples) and the validation set (5k examples). We train the model for 4 epochs with a 2e-5 learning rate and a 16- batch size, using the Adam optimizer. During the evaluation, we consider an example coherent when the probability of being coherent predicted by the classifier is greater than 0.5. We use the ratio of outputs (along with the input) that are classified as coherent by the classifier to all generated outputs as the coherence score. The result of the automatic evaluation is presented in Table 3. According to the table, compared to the BART baseline, our model consistently generates more word overlaps with ground truth and achieves better diversity under the guidance of character representations, which means our model can generate more diverse but not commonplace responses. Figure 3 plots the coherence score varying with the number of masked dialogue turns. The result shows that our model gets a higher coherence score than BART when required to generate more than seven turns of dialogue in one story. Manual Evaluation We conduct a pairwise comparison between our model and the BART baseline. We randomly select 100 examples from the test set. For each pair of outputs along with the input, we ask three annotators to give a preference (win, lose and tie) in terms of fluency, coherence, and informativeness. All the annotations are native Chinese speakers. We adopt majority voting to make final decisions among the annotators. The three aspects of manual evaluation are as follows: Fluency: Grammatical correctness and intra-sentence linguistic quality.


Coherence: Inter-sentence relatedness, causal and temporal dependencies. We judge the coherence between the story and a dialogue turn by following the criterion in Table 5. We add the scores of all the generated dialogue turns in a story to get the overall coherence score of the story, which is then used to compare with each other.


Informativeness: Interesting, diverse and rich details. As shown in Table 4, all the results show moderate (κ > 0.4) agreement, which shows our model outperforms the BART baseline significantly in dialogue informativeness and coherence. Case Study Figure 5 showed two examples to investigate how learning character representations can help our model generate more coherent dialogue. We found that our model can better model the relationship between different characters and the direction of the storyline. For example, in the first case, we can see that the BART’s generation confuses different characters’ fathers, while our model captures the relationship between different characters, and generates proper responses for the corresponding characters, which also moves the plot forward. And in the second case, we can see that BART’s generation is commonplace and contradicts the plot development. In contrast, our model captures the intentions of the speaker and the development trend of the plot, generating an appropriate and coherent response. Since these two models use the same pretrained weight, we can infer that the character modeling module leverages the coherent and reasonable generation. We also summarize four error types of the generated dialogue turn for the DialGen task: (1) Intersentence Contradiction; (2) Inter-sentence Repetition; (3) Intra-sentence Contradiction; (4) Intrasentence Repetition. We show the typical corresponding cases in Figure 4. We conducted a quantitative analysis of those 4 error types on our model’s generation. We analyzed 20 stories with 103 dialog turns and the results are shown in Figure 6. We found that both our model and BART suffer from these errors, suggesting that there is still space for model improvement, especially in the inter-sentence repetition. 6.2 Dialogue Speaker Recognition Implementation Details To conduct experiments on the speaker recognition task, we decide the hyper-parameters based on the performance of the validation set. For the BART baseline and our approach, we insert a mask token before each dialogue needed to be predicted, and a person id token before and after each character name span. Then, we insert all the unique person id tokens before the input stories as different options, and make predictions based on the cos similarity of option tokens and mask tokens. We train Shao et al. (2021)’s BART model for 30 epochs with a 5e-5 learning rate for 3 days, For encoder-only baselines, we implemented BERT, RoBERTa, and MacBERT and trained them for 15 epochs wity a le-5 learning rate for 2 days. For our model, we train it for 22 epochs with a le-6 learning rate for 2 days. All baselines and our model are trained using the Adam optimizer. Metrics We evaluate the DialSpk task using two automatic metrics including dialogue-level accuracy (DAC) and story-level accuracy (SAC). DAC is calculated as the ratio of the correct predictions to the total number of specifies dialogue turns, while SAC is the ratio of the number of stories where all dialogue turns are correctly predicted to the number of all test examples. These two metrics provide the evaluation for dialogue understanding with different granularities. Results As shown in Table 7, our model outperforms all the baselines significantly (p< 0.01, Wilcoxon signed-rank test) on both DAS and SAC scores, suggesting the benefit of learning character representations. We tested the accuracy of automatic training set annotations, and the DAC/SAC scores are 86.78%/67.80%. Together with the model’s performance on the test set, we can see the automatic annotation for the training set is of good quality. We also conducted the human prediction experiment, and the DAC/SAC scores are 97.90%/90.70%, which are much higher than the best model. So there is much room for further improvement for machine-based approaches. This paper is available on arxiv under CC 4.0 DEED license. [2] https://huggingface.co/fnlp/ bart-base-chinese [3] https://huggingface.co/bert-base-chinese [4] https://huggingface.co/hfl/ chinese-roberta-wwm-ext [5] https://huggingface.co/hfl/ chinese-macbert-base Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 6 Experiments 6.1 Masked Dialogue Generation Implementation Details To conduct experiments on the masked dialogue generation task, we decide the hyper-parameters based on the performance of the validation set. We train Shao et al. (2021)’s BART model for 4.6 epochs with a 1e-4 learning rate for 1 day, and for our model, we train it for 5.6 epochs with a 1e-4 learning rate for 1 day. All baselines and our model are trained using the Adam optimizer. Implementation Details During the training process for our method, we computed the selection coverage of characters within a single story. And it showed that in every 1000 training steps, the coverage of different characters ranged from 98.64% to 99.00%, which meant nearly all the characters are selected during training, and all the characters contributed to the generated dialogue. It further proved that the argmax in Eq. 3 operation doesn’t break the gradient progress when training for this task. Automatic Evaluation Following previous works, we use several standard, widely used automatic evaluation metrics. We use BLEUn (Papineni et al., 2002) to measure the average word overlap between each generated and groundtruth dialogue turn (n=1,2), and Distinct-n (Li et al., 2015) to evaluate n-gram diversity of generated dialogue turns (n=2,3,4). Automatic Evaluation To be more specific, for the coherence classifier, we construct the training and validation sets by randomly shuffling the order of dialogue turns and keeping other content in the correct order. We regard the perturbed story as a negative example and the original story as a positive example. We sample another 195k stories (except those in DIALSTORY) from the novels of Guan et al. (2022) to construct the training set (190k examples) and the validation set (5k examples). We train the model for 4 epochs with a 2e-5 learning rate and a 16- batch size, using the Adam optimizer. During the evaluation, we consider an example coherent when the probability of being coherent predicted by the classifier is greater than 0.5. We use the ratio of outputs (along with the input) that are classified as coherent by the classifier to all generated outputs as the coherence score. The result of the automatic evaluation is presented in Table 3. According to the table, compared to the BART baseline, our model consistently generates more word overlaps with ground truth and achieves better diversity under the guidance of character representations, which means our model can generate more diverse but not commonplace responses. Figure 3 plots the coherence score varying with the number of masked dialogue turns. The result shows that our model gets a higher coherence score than BART when required to generate more than seven turns of dialogue in one story. Manual Evaluation We conduct a pairwise comparison between our model and the BART baseline. We randomly select 100 examples from the test set. For each pair of outputs along with the input, we ask three annotators to give a preference (win, lose and tie) in terms of fluency, coherence, and informativeness. All the annotations are native Chinese speakers. We adopt majority voting to make final decisions among the annotators. The three aspects of manual evaluation are as follows: Manual Evaluation Fluency: Grammatical correctness and intra-sentence linguistic quality. Coherence: Inter-sentence relatedness, causal and temporal dependencies. We judge the coherence between the story and a dialogue turn by following the criterion in Table 5. We add the scores of all the generated dialogue turns in a story to get the overall coherence score of the story, which is then used to compare with each other. Informativeness: Interesting, diverse and rich details. Fluency: Grammatical correctness and intra-sentence linguistic quality. Fluency : Grammatical correctness and intra-sentence linguistic quality. Fluency Coherence: Inter-sentence relatedness, causal and temporal dependencies. We judge the coherence between the story and a dialogue turn by following the criterion in Table 5. We add the scores of all the generated dialogue turns in a story to get the overall coherence score of the story, which is then used to compare with each other. Coherence : Inter-sentence relatedness, causal and temporal dependencies. We judge the coherence between the story and a dialogue turn by following the criterion in Table 5. We add the scores of all the generated dialogue turns in a story to get the overall coherence score of the story, which is then used to compare with each other. Coherence Informativeness: Interesting, diverse and rich details. Informativeness : Interesting, diverse and rich details. Informativeness As shown in Table 4, all the results show moderate (κ > 0.4) agreement, which shows our model outperforms the BART baseline significantly in dialogue informativeness and coherence. Case Study Figure 5 showed two examples to investigate how learning character representations can help our model generate more coherent dialogue. We found that our model can better model the relationship between different characters and the direction of the storyline. For example, in the first case, we can see that the BART’s generation confuses different characters’ fathers, while our model captures the relationship between different characters, and generates proper responses for the corresponding characters, which also moves the plot forward. And in the second case, we can see that BART’s generation is commonplace and contradicts the plot development. In contrast, our model captures the intentions of the speaker and the development trend of the plot, generating an appropriate and coherent response. Since these two models use the same pretrained weight, we can infer that the character modeling module leverages the coherent and reasonable generation. Case Study We also summarize four error types of the generated dialogue turn for the DialGen task: (1) Intersentence Contradiction; (2) Inter-sentence Repetition; (3) Intra-sentence Contradiction; (4) Intrasentence Repetition. We show the typical corresponding cases in Figure 4. We conducted a quantitative analysis of those 4 error types on our model’s generation. We analyzed 20 stories with 103 dialog turns and the results are shown in Figure 6. We found that both our model and BART suffer from these errors, suggesting that there is still space for model improvement, especially in the inter-sentence repetition. 6.2 Dialogue Speaker Recognition Implementation Details To conduct experiments on the speaker recognition task, we decide the hyper-parameters based on the performance of the validation set. For the BART baseline and our approach, we insert a mask token before each dialogue needed to be predicted, and a person id token before and after each character name span. Then, we insert all the unique person id tokens before the input stories as different options, and make predictions based on the cos similarity of option tokens and mask tokens. We train Shao et al. (2021)’s BART model for 30 epochs with a 5e-5 learning rate for 3 days, For encoder-only baselines, we implemented BERT, RoBERTa, and MacBERT and trained them for 15 epochs wity a le-5 learning rate for 2 days. For our model, we train it for 22 epochs with a le-6 learning rate for 2 days. All baselines and our model are trained using the Adam optimizer. Metrics We evaluate the DialSpk task using two automatic metrics including dialogue-level accuracy (DAC) and story-level accuracy (SAC). DAC is calculated as the ratio of the correct predictions to the total number of specifies dialogue turns, while SAC is the ratio of the number of stories where all dialogue turns are correctly predicted to the number of all test examples. These two metrics provide the evaluation for dialogue understanding with different granularities. Metrics Results As shown in Table 7, our model outperforms all the baselines significantly (p< 0.01, Wilcoxon signed-rank test) on both DAS and SAC scores, suggesting the benefit of learning character representations. We tested the accuracy of automatic training set annotations, and the DAC/SAC scores are 86.78%/67.80%. Together with the model’s performance on the test set, we can see the automatic annotation for the training set is of good quality. We also conducted the human prediction experiment, and the DAC/SAC scores are 97.90%/90.70%, which are much higher than the best model. So there is much room for further improvement for machine-based approaches. Results This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv [2] https://huggingface.co/fnlp/ bart-base-chinese [3] https://huggingface.co/bert-base-chinese [4] https://huggingface.co/hfl/ chinese-roberta-wwm-ext [5] https://huggingface.co/hfl/ chinese-macbert-base

Understanding and Generating Dialogue between Characters in Stories: Experiments

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 5 Methodology We propose to learn representations of different characters and exert them on decoding masked dialogue turns or predicting speakers. In this section, we describe the details of our model. Figure 2 shows the model overview for the DialGen task. 5.1 Character Representation Learning 5.2 Character Representation Utilization This paper is available on arxiv under CC 4.0 DEED license. Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 5 Methodology We propose to learn representations of different characters and exert them on decoding masked dialogue turns or predicting speakers. In this section, we describe the details of our model. Figure 2 shows the model overview for the DialGen task. 5.1 Character Representation Learning 5.2 Character Representation Utilization This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv

Understanding and Generating Dialogue between Characters in Stories: Methodology

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 4 Proposed Tasks We aim to measure model’s ability to understand and generate dialogue in a story. To this end, we design the dialogue generation task Masked Dialogue Generation and dialogue understanding task Dialogue Speaker Recognition. We show the task definitions, targets, dataset construction and statistics below. 4.1 Masked Dialogue Generation Dataset Construction We use the following constraints to construct the DialGen dataset based on DIALSTORY: • We randomly mask 30% of the dialogue turns in each story. • We do not mask the first 50 tokens to provide sufficient background information for the story. • We do not mask the last 30 tokens to provide ending information for that story. • We ensure that each input story (i.e. with masked dialogue turns) mentions at least five characters. Table 2 shows the detailed statistics. 4.2 Dialogue Speaker Recognition Dataset Construction We randomly sampled 20k stories from DIALSTORY and automatically annotate the speaker for each dialogue turn for training, and resorted to manual annotation for validation and testing. For manual annotation, we first ask one annotator to label the characters in a story and the speaker of each dialogue turn. Then we asked another two annotators to check the correctness of the annotations, e.g., whether all mentioned characters are annotated, and whether each dialogue speaker is correct. We require the first annotator to re-annotate those examples that another two annotators do not agree on, and repeat the above process until all annotators agree on the examples. We also sampled 100 stories in the training set for manual annotation to investigate the accuracy of automatic annotation, which we will discuss in Section 6.2. Table 2 shows the detailed statistics. This paper is available on arxiv under CC 4.0 DEED license. Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 4 Proposed Tasks We aim to measure model’s ability to understand and generate dialogue in a story. To this end, we design the dialogue generation task Masked Dialogue Generation and dialogue understanding task Dialogue Speaker Recognition. We show the task definitions, targets, dataset construction and statistics below. 4.1 Masked Dialogue Generation Dataset Construction We use the following constraints to construct the DialGen dataset based on DIALSTORY: Dataset Construction • We randomly mask 30% of the dialogue turns in each story. • We do not mask the first 50 tokens to provide sufficient background information for the story. • We do not mask the last 30 tokens to provide ending information for that story. • We ensure that each input story (i.e. with masked dialogue turns) mentions at least five characters. Table 2 shows the detailed statistics. 4.2 Dialogue Speaker Recognition Dataset Construction We randomly sampled 20k stories from DIALSTORY and automatically annotate the speaker for each dialogue turn for training, and resorted to manual annotation for validation and testing. For manual annotation, we first ask one annotator to label the characters in a story and the speaker of each dialogue turn. Then we asked another two annotators to check the correctness of the annotations, e.g., whether all mentioned characters are annotated, and whether each dialogue speaker is correct. We require the first annotator to re-annotate those examples that another two annotators do not agree on, and repeat the above process until all annotators agree on the examples. We also sampled 100 stories in the training set for manual annotation to investigate the accuracy of automatic annotation, which we will discuss in Section 6.2. Table 2 shows the detailed statistics. Dataset Construction This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv

Understanding and Generating Dialogue between Characters in Stories: Proposed Tasks

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 3 DIALSTORY Dataset We construct the DIALSTORY dataset by randomly sampling 105k chapters from the Chinese novels released by Guan et al. (2022) with each chapter including at least ten dialogue turns. We also set a restriction that the number of tokens in all dialogue turns should account for at least 30% and at most 50% of the total length of the story, in order to keep a balance between the context and dialogue. We automatically annotate dialogue turns in these stories as text spans that are surrounded by quotation marks. Then, we use a pretrained named entity recognition model (Zhao et al., 2019) to identify all people’s names. Each distinct name corresponds to a character. We also conduct a manual annotation on 150 stories, and the accuracy of character identification is 718/746=96.2%, which shows the high quality of this automatic method. We then decide the speaker of the dialogue by recognizing the subjects of sentences before and after the dialogue turn using spaCy[1]. Table 1 shows the statistics of our dataset. This paper is available on arxiv under CC 4.0 DEED license. [1] https://spacy.io/ Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 3 DIALSTORY Dataset We construct the DIALSTORY dataset by randomly sampling 105k chapters from the Chinese novels released by Guan et al. (2022) with each chapter including at least ten dialogue turns. We also set a restriction that the number of tokens in all dialogue turns should account for at least 30% and at most 50% of the total length of the story, in order to keep a balance between the context and dialogue. We automatically annotate dialogue turns in these stories as text spans that are surrounded by quotation marks. Then, we use a pretrained named entity recognition model (Zhao et al., 2019) to identify all people’s names. Each distinct name corresponds to a character. We also conduct a manual annotation on 150 stories, and the accuracy of character identification is 718/746=96.2%, which shows the high quality of this automatic method. We then decide the speaker of the dialogue by recognizing the subjects of sentences before and after the dialogue turn using spaCy[1]. Table 1 shows the statistics of our dataset. This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv [1] https://spacy.io/

Understanding and Generating Dialogue between Characters in Stories: DIALSTORY Dataset

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References 2 Related Works Story Understanding and Generation Recently there have been various tasks proposed to assess the ability of story understanding and generation such as story ending selection (Zhou et al., 2019), story ending generation (Guan et al., 2019), story completion (Wang and Wan, 2019), story character identification (Brahman et al., 2021), story generation from prompts (Fan et al., 2018), titles (Yao et al., 2019) and beginnings (Guan et al., 2020). Another line of work focused on controllable attributes in story generation such as keywords (Xu et al., 2020), outlines (Rashkin et al., 2020), emotional trajectories (Brahman and Chaturvedi, 2020), styles (Kong et al., 2021) and characters’ personalities (Zhang et al., 2022). Different from them, we emphasize the importance of dialogue in developing characters and maintaining the story’s coherence. Dialogue in story The dialogue context is comprised of a variety of messages from users and the system, determining the conversation topic and the user’s goal for the conversation(Serban et al., 2017). The dialogue in the story makes this feature more apparent. Recently, like AI Dungeon, some text adventure games focus on players and machines co-creating stories. Li et al. (2022) proposed the idea that players can play the role of the chosen character and chat with others through the dialogue, while Xi et al. (2021) presented an AI game where players can explore different ways to reach the plot goals. In contrast, our tasks focus on both the understanding and generation of dialogue in the story, and they can be easily generalized to more complex AI interactive games. Character Representations Dialogue unfolds in a story around several inter-related characters. Accordingly, it is essential to capture the traits of different characters for dialogue modeling. Ji et al. (2017); Clark et al. (2018) proposed to update characters’ representations dynamically based on previous hidden outputs to track their states, which does not apply to the parallel architecture of Transformer for training. Azab et al. (2019) derived character representations from their corresponding dialogue turns. In contrast, we learn character representations from story plots, which then serve to understand and generate dialogue. Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References 2 Related Works Story Understanding and Generation Recently there have been various tasks proposed to assess the ability of story understanding and generation such as story ending selection (Zhou et al., 2019), story ending generation (Guan et al., 2019), story completion (Wang and Wan, 2019), story character identification (Brahman et al., 2021), story generation from prompts (Fan et al., 2018), titles (Yao et al., 2019) and beginnings (Guan et al., 2020). Another line of work focused on controllable attributes in story generation such as keywords (Xu et al., 2020), outlines (Rashkin et al., 2020), emotional trajectories (Brahman and Chaturvedi, 2020), styles (Kong et al., 2021) and characters’ personalities (Zhang et al., 2022). Different from them, we emphasize the importance of dialogue in developing characters and maintaining the story’s coherence. Story Understanding and Generation Dialogue in story The dialogue context is comprised of a variety of messages from users and the system, determining the conversation topic and the user’s goal for the conversation(Serban et al., 2017). The dialogue in the story makes this feature more apparent. Recently, like AI Dungeon, some text adventure games focus on players and machines co-creating stories. Li et al. (2022) proposed the idea that players can play the role of the chosen character and chat with others through the dialogue, while Xi et al. (2021) presented an AI game where players can explore different ways to reach the plot goals. In contrast, our tasks focus on both the understanding and generation of dialogue in the story, and they can be easily generalized to more complex AI interactive games. Dialogue in story Character Representations Dialogue unfolds in a story around several inter-related characters. Accordingly, it is essential to capture the traits of different characters for dialogue modeling. Ji et al. (2017); Clark et al. (2018) proposed to update characters’ representations dynamically based on previous hidden outputs to track their states, which does not apply to the parallel architecture of Transformer for training. Azab et al. (2019) derived character representations from their corresponding dialogue turns. In contrast, we learn character representations from story plots, which then serve to understand and generate dialogue. Character Representations

Understanding and Generating Dialogue between Characters in Stories: Related Works

Authors:
(1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology;
(4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Related Works DIALSTORY Dataset Proposed Tasks Methodology Experiments Discussion Future Work Conclusion Limitations and References Abstract Many classical fairy tales, fictions, and screenplays leverage dialogue to advance story plots and establish characters. We present the first study to explore whether machines can understand and generate dialogue in stories, which require capturing traits of different characters and the relationships between them. To this end, we propose two new tasks including Masked Dialogue Generation and Dialogue Speaker Recognition, i.e., generating missing dialogue turns and predicting speakers for specified dialogue turns, respectively. We build a new dataset DIALSTORY, which consists of 105k Chinese stories with a large amount of dialogue weaved into the plots to support the evaluation. We show the difficulty of the proposed tasks by testing existing models with automatic and manual evaluation on DIALSTORY. Furthermore, we propose to learn explicit character representations to improve performance on these tasks. Extensive experiments and case studies show that our approach can generate more coherent and informative dialogue, and achieve higher speaker recognition accuracy than strong baselines. 1 Introduction Dialogue plays an important role in various types of literary works such as short stories, novels, and screenplays by advancing plots, establishing characters, and providing expositions using natural and lifelike words (Kennedy et al., 1983). Compared to dialogue in conversational scenarios such as chitchat bots (Shang et al., 2015) or task-oriented dialogue systems (Deng et al., 2012), dialogue in stories is mainly used to exhibit emotions, motivations, or personalities of characters following the authors’ design, which further serve for the coherence, informativeness, engagingness, and plot development of whole stories. Dialogue is also revealed to be essential for user-agent interaction in many text adventure games (Xi et al., 2021;Li et al., 2022). It has not been widely explored for machines to understand and generate dialogue in stories despite the broad recognition of the importance of this ability. To spur research in this field, we present a new story dataset named DIALSTORY, which consists of 105k Chinese short stories with automatic annotations of dialogue turns and corresponding speakers. Furthermore, we formulate two new tasks including: (1) Masked Dialogue Generation (DialGen): completing a story with several dialogue turns masked with placeholders; and (2) Dialogue Speaker Recognition (DialSpk): choosing correct speakers from given candidates for several specified dialogue turns. We construct standardized datasets for these tasks through automatic or manual annotation based on DIALSTORY. As exemplified in Figure 1, these tasks comprehensively investigate the ability to capture characters’ emotions (e.g., the mother squirrel is worried about baby squirrel’s catching cold), motivations (e.g., the mother squirrel intends to call her baby to come back home or get medicine), and relationship between each other (e.g. mother rabbit knows the kind of flower as a friend of mother squirrel when they were kids). Furthermore, we found in massive stories that dialogue had a strong connection to different characters, reflected in their emotional states, speaking styles, and plot development. To provide more insights to tackle these tasks, we propose to learn character representations for modeling dependencies between characters and dialogue explicitly. Extensive experiments and case studies show that our approach can generate more coherent and informative dialogue, and achieve higher speaker recognition accuracy than strong baselines. This paper is available on arxiv under CC 4.0 DEED license. Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Authors: Authors: (1) Jianzhu Yao, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (2) Ziqi Liu, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (3) Jian Guan, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology; (4) Minlie Huang, The CoAI group, Tsinghua University, Beijing, China Department of Computer Science and Technology, Tsinghua University, Beijing, China Beijing National Research Center for Information Science and Technology. Table of Links Abstract and Intro Abstract and Intro Related Works Related Works DIALSTORY Dataset DIALSTORY Dataset Proposed Tasks Proposed Tasks Methodology Methodology Experiments Experiments Discussion Discussion Future Work Future Work Conclusion Conclusion Limitations and References Limitations and References Abstract Many classical fairy tales, fictions, and screenplays leverage dialogue to advance story plots and establish characters. We present the first study to explore whether machines can understand and generate dialogue in stories, which require capturing traits of different characters and the relationships between them. To this end, we propose two new tasks including Masked Dialogue Generation and Dialogue Speaker Recognition, i.e., generating missing dialogue turns and predicting speakers for specified dialogue turns, respectively. We build a new dataset DIALSTORY, which consists of 105k Chinese stories with a large amount of dialogue weaved into the plots to support the evaluation. We show the difficulty of the proposed tasks by testing existing models with automatic and manual evaluation on DIALSTORY. Furthermore, we propose to learn explicit character representations to improve performance on these tasks. Extensive experiments and case studies show that our approach can generate more coherent and informative dialogue, and achieve higher speaker recognition accuracy than strong baselines. 1 Introduction Dialogue plays an important role in various types of literary works such as short stories, novels, and screenplays by advancing plots, establishing characters, and providing expositions using natural and lifelike words (Kennedy et al., 1983). Compared to dialogue in conversational scenarios such as chitchat bots (Shang et al., 2015) or task-oriented dialogue systems (Deng et al., 2012), dialogue in stories is mainly used to exhibit emotions, motivations, or personalities of characters following the authors’ design, which further serve for the coherence, informativeness, engagingness, and plot development of whole stories. Dialogue is also revealed to be essential for user-agent interaction in many text adventure games (Xi et al., 2021;Li et al., 2022). It has not been widely explored for machines to understand and generate dialogue in stories despite the broad recognition of the importance of this ability. To spur research in this field, we present a new story dataset named DIALSTORY, which consists of 105k Chinese short stories with automatic annotations of dialogue turns and corresponding speakers. Furthermore, we formulate two new tasks including: (1) Masked Dialogue Generation (DialGen): completing a story with several dialogue turns masked with placeholders; and (2) Dialogue Speaker Recognition (DialSpk): choosing correct speakers from given candidates for several specified dialogue turns. We construct standardized datasets for these tasks through automatic or manual annotation based on DIALSTORY. As exemplified in Figure 1, these tasks comprehensively investigate the ability to capture characters’ emotions (e.g., the mother squirrel is worried about baby squirrel’s catching cold), motivations (e.g., the mother squirrel intends to call her baby to come back home or get medicine), and relationship between each other (e.g. mother rabbit knows the kind of flower as a friend of mother squirrel when they were kids). Furthermore, we found in massive stories that dialogue had a strong connection to different characters, reflected in their emotional states, speaking styles, and plot development. To provide more insights to tackle these tasks, we propose to learn character representations for modeling dependencies between characters and dialogue explicitly. Extensive experiments and case studies show that our approach can generate more coherent and informative dialogue, and achieve higher speaker recognition accuracy than strong baselines. This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv

Understanding and Generating Dialogue between Characters in Stories: Abstract and Intro

Storing Passwords In A Database: A Better Way to Do So

‘ai tools’ Image created by HackerNoon AI Image Generator Hey Hacker, Unless you’ve been living under a rock, you know that AI tools have been making some big waves, flooding the digital works with AI-generated content. Whether you’re an AI supporter or hater, you can’t deny it’s been helping us see the world from a different angle! Let’s cut to the chase: as OpenAI announces the arrival of GPT-4 and boasts about its ability to problem solve with greater accuracy, the HackerNoon community gets busy chatting about all things AI, such as programming tools, text processors, and AI applicationsin general. We're such nerds that we've even added a fancy new AI content detector to our editor🧐. As always, we’re excited to hear YOUR opinions on these topics. We know, writer’s block can catch even the best of us. So we put together some idea prompts to get you started on your next opinion piece: Will AI tools change the way educators evaluate student work? Do you think AI tools are ready to be used as educational resources? Share your thoughts here.


How do you think the big streamer services like YouTube and Netflix will be impacted by the new generative video tools? Write about it here.


AI-generated visual art often has a distinguishable appearance - will the art world designate a new classification to this type of art form? Tell us about it using this template.


Is Chat-GPT really gonna help you become a better programmer? Is AI even good at web development, or are people just delusional? Spill the tea here!


What new AI tools are you inspired by? Tell us about it here.
We can see the wheels turning already - click one of the links above and get writing!
Some Added Writing Tips to Make Your Stories Stand Out Like Skinny Jeans on a Millenial:

Check out the active writing contests and tag your story to win prizes 🥇
You can use AI tools, but disclose which sections were written this way 🦾

We KNOW you’re going to serve us some juicy gossip on these new technologies - can’t wait to spread it all over HackerNoon’s homepage!Happy writing,The HackerNoon Editorial Team ‘ai tools’ Image created by HackerNoon AI Image Generator ‘ai tools’ HackerNoon AI Image Generator Hey Hacker, Unless you’ve been living under a rock, you know that AI tools have been making some big waves, flooding the digital works with AI-generated content. Whether you’re an AI supporter or hater, you can’t deny it’s been helping us see the world from a different angle! Let’s cut to the chase: as OpenAI announces the arrival of GPT-4 and boasts about its ability to problem solve with greater accuracy, the HackerNoon community gets busy chatting about all things AI , such as programming tools , text processors , and AI applications in general. GPT-4 all things AI programming tools text processors AI applications We're such nerds that we've even added a fancy new AI content detector to our editor🧐. new AI content detector As always, we’re excited to hear YOUR opinions on these topics. We know, writer’s block can catch even the best of us. So we put together some idea prompts to get you started on your next opinion piece: As always, we’re excited to hear YOUR opinions on these topics. We know, writer’s block can catch even the best of us. So we put together some idea prompts to get you started on your next opinion piece: YOUR opinions Will AI tools change the way educators evaluate student work? Do you think AI tools are ready to be used as educational resources? Share your thoughts here. How do you think the big streamer services like YouTube and Netflix will be impacted by the new generative video tools? Write about it here. AI-generated visual art often has a distinguishable appearance - will the art world designate a new classification to this type of art form? Tell us about it using this template. Is Chat-GPT really gonna help you become a better programmer? Is AI even good at web development, or are people just delusional? Spill the tea here! What new AI tools are you inspired by? Tell us about it here.
We can see the wheels turning already - click one of the links above and get writing!
Some Added Writing Tips to Make Your Stories Stand Out Like Skinny Jeans on a Millenial:

Check out the active writing contests and tag your story to win prizes 🥇
You can use AI tools, but disclose which sections were written this way 🦾

We KNOW you’re going to serve us some juicy gossip on these new technologies - can’t wait to spread it all over HackerNoon’s homepage!Happy writing,The HackerNoon Editorial Team Will AI tools change the way educators evaluate student work? Do you think AI tools are ready to be used as educational resources? Share your thoughts here. Will AI tools change the way educators evaluate student work? Do you think AI tools are ready to be used as educational resources? Share your thoughts here . here How do you think the big streamer services like YouTube and Netflix will be impacted by the new generative video tools? Write about it here. How do you think the big streamer services like YouTube and Netflix will be impacted by the new generative video tools? Write about it here . here AI-generated visual art often has a distinguishable appearance - will the art world designate a new classification to this type of art form? Tell us about it using this template. AI-generated visual art often has a distinguishable appearance - will the art world designate a new classification to this type of art form? Tell us about it using this template . template Is Chat-GPT really gonna help you become a better programmer? Is AI even good at web development, or are people just delusional? Spill the tea here! Is Chat-GPT really gonna help you become a better programmer? Is AI even good at web development, or are people just delusional? Spill the tea here ! here What new AI tools are you inspired by? Tell us about it here. We can see the wheels turning already - click one of the links above and get writing! Some Added Writing Tips to Make Your Stories Stand Out Like Skinny Jeans on a Millenial: Check out the active writing contests and tag your story to win prizes 🥇
You can use AI tools, but disclose which sections were written this way 🦾 We KNOW you’re going to serve us some juicy gossip on these new technologies - can’t wait to spread it all over HackerNoon’s homepage! Happy writing, The HackerNoon Editorial Team What new AI tools are you inspired by? Tell us about it here . here We can see the wheels turning already - click one of the links above and get writing! Some Added Writing Tips to Make Your Stories Stand Out Like Skinny Jeans on a Millenial: Check out the active writing contests and tag your story to win prizes 🥇 You can use AI tools, but disclose which sections were written this way 🦾 Check out the active writing contests and tag your story to win prizes 🥇 writing contests You can use AI tools, but disclose which sections were written this way 🦾 We KNOW you’re going to serve us some juicy gossip on these new technologies - can’t wait to spread it all over HackerNoon’s homepage! Happy writing, The HackerNoon Editorial Team

Stories, News & Mentions about STORI

Stori’s Stories on HackerNoon