Use Case
The Secret Behind Corpuschat: Discipline-specific Corpora
CorpusChat is a total of six discipline-specific chatbots created using academic texts across many disciplinesm, supported by the Teaching and Development Grant. CorpusChat integrates Corpus Linguistics and GenAI which are two data-centric domains with a synergistic relationship, to help mitigate common concerns GenAI poses such as accuracy, reliability and hallucinations. This is done by restricting the AI to specific academic corpora and verified sources, preventing the generic, robotic prose of public LLMs.
Use Case
The Secret Behind Corpuschat: Discipline-specific Corpora
About
CorpusChat is a total of six discipline-specific chatbots created using academic texts across many disciplinesm, supported by the Teaching and Development Grant. CorpusChat integrates Corpus Linguistics and GenAI which are two data-centric domains with a synergistic relationship, to help mitigate common concerns GenAI poses such as accuracy, reliability and hallucinations. This is done by restricting the AI to specific academic corpora and verified sources, preventing the generic, robotic prose of public LLMs.
Connect to the Contributor
Dr. Lisa Cheung
Senior Lecturer
Centre for Applied English Studies, Faculty of Arts
Faculty / Unit
Faculty of Arts
Professional Development Event
Harnessing GenAI for T&L sharing
April 16, 2025
AI4TL Conference
December 3, 2025
9th HKU Teaching and Learning Festival
May 26, 2026
AI LiteracyArtificial Intelligence for Academic Purposes (AIAP)Contextual Knowledge BaseCorpus LinguisticsCorpus-based ChatbotCorpuschatData-Driven Learning (DDL)Disciplinary writingDiscipline SpecificDiscipline-specific vocabularyEthical ConcernsGenAIGenerative Pre-training Transformer (GPT)Learning AssistantRetrieval Augmented Generation (RAG)Retrieval-augmented generation (RAG)Sociocultural theoryStudent FeedbackStudent-centered pedagogyWord collocationsWritingWriting Coach
AI LiteracyArtificial Intelligence for Academic Purposes (AIAP)Contextual Knowledge BaseCorpus LinguisticsCorpus-based ChatbotCorpuschatData-Driven Learning (DDL)Disciplinary writingDiscipline SpecificDiscipline-specific vocabularyEthical ConcernsGenAIGenerative Pre-training Transformer (GPT)Learning AssistantRetrieval Augmented Generation (RAG)Retrieval-augmented generation (RAG)Sociocultural theoryStudent FeedbackStudent-centered pedagogyWord collocationsWritingWriting Coach