ChatGPT-Based Learning And Reading Assistant (C-LARA): Second Report

Résumé 0

ChatGPT-based Learning And Reading Assistant (C-LARA – pronounced “Clara”) is an AIbasedplatform which allows users to create multimodal texts designed to improve reading skillsin second languages. GPT-4/ChatGPT-4 is central to the project: as well as being the corelanguage processing component, it has in collaboration with a human partner developed thegreater part of the codebase.Following on from the initial progress report, released in July 2023, we focus on new workcarried out during the period August 2023 – March 2024. The platform is far more usable. CLARAis now packaged with a wizard-style interface (“Simple C-LARA”) that allows the nonexpertuser to create a complete illustrated multimodal text by entering a prompt and approvingdefault choices a few times, and the software is deployed on a fast dedicated server maintained bythe University of South Australia. Other substantial new pieces of functionality are support for“phonetic texts”, where words are automatically divided up into units associated with phoneticvalues; “reading histories”, which support the combination of several texts into a single virtualdocument; and the social network, rudimentary in the first version, which now includes supportfor friending, an update feed, and email alerts.To investigate the AI’s abilities as a language processor, we present an experiment where wecreated six texts for each of five languages, using the same prompts for each language, andevaluated the accuracy of the language processing. We also give the results when some of theexperiments were repeated five months later with a newer version of GPT-4, in the case ofEnglish revealing a dramatic reduction in error rates. A small questionnaire-based study probesusers’ subjective views of C-LARA projects they have created: in general, people are pleasedwith the results, to the extent that they are often sharing them.With regard to GPT-4/ChatGPT-4’s software engineer role, we present a breakdown of thevarious modules and functionalities, indicating the AI’s contribution. It is capable of writing thesimpler modules on its own or with minimal human assistance, and only had serious problemswith a small number of top-level functionalities, in particular “Simple C-LARA”, which directlyor indirectly involved most of the codebase.We describe initial use cases, including trialling of C-LARA in a school classroom, integratingit into the experimental CALL platform Basm, and creating multimodal texts in the Oceaniclanguages Drehu and Iaai. A short section summarises our policy on ethical issues concerningthe crediting of the AI as an author. The appendices present examples illustrating use of theSimple C-LARA and Advanced C-LARA versions of the platform, list functionalities and codefiles, and reproduce conversations with the AI about various aspects of the project.