Information and Updates
- Location: NeurIPS'23 Workshop, New Orleans, Louisiana, USA.
- Date: 15 December 2023 (Friday).
- Room: 265—268.
- Schedule: See more information below in Schedule.
- Contact: For any questions, please email contact.gaied AT gmail.com
- Updates:
- 2024-02-04: An overview article about the GAIED workshop is publicly available here!
- 2024-01-20: Video recordings of talks and the panel are publicly available here!
- 2023-12-05: We have added details about the schedule!
- 2023-12-05: We have added details about the papers!
- 2023-11-01: We have updated the list of invited speakers!
- 2023-08-05: We have added details about paper submissions!
- 2023-07-05: Initial website created for the NeurIPS'23 workshop!
- Please refresh the page to ensure you have the latest content!
Overview
GAIED (pronounced "guide") aims to bring together researchers, educators, and practitioners to explore the potential of generative AI for enhancing education. Such an exploration, jointly as a community, is time critical: Recent advances in generative AI, in particular deep generative and large language models like ChatGPT, are bringing in transformational effects on the educational landscape. On the one hand, these advances provide unprecedented opportunities to enhance education by creating unique human-machine collaborative systems, e.g., these models could act as personalized digital tutors for students, as digital assistants for educators, and as digital peers to enable new collaborative learning scenarios. On the other hand, the advanced capabilities of these generative AI models have brought unexpected challenges for educators and policymakers worldwide, causing a chaotic disruption in universities and schools to design regulatory policies about the usage of these models. The workshop will investigate these opportunities and challenges in education by focusing the discussions along two thrusts:
- GAI→ED: Exploring how recent advances in generative AI provide new opportunities to drastically improve state-of-the-art educational technology.
- ED→GAI: Identifying unique challenges in education caused by these recent advances and how to tackle them by bringing in desired safeguards along with technical innovations in generative AI.
For us to fully realize these opportunities and tackle these challenges, it is crucial to build a community of researchers, educators, and practitioners that are "multilingual" with (a) technical expertise in the cutting-edge advances in generative AI, (b) first-hand experience of working with students in classrooms, and (c) know-how of building/deploying educational technology at scale. The goal of GAIED is to foster such a multilingual community. The workshop will bring together speakers and participants with diverse backgrounds ranging from researchers in human-computer interaction, learning sciences, natural language processing, and program synthesis to industry practitioners and educators directly involved in educational activities. Moreover, the workshop program, featuring diverse speakers and panelists, is designed to facilitate new connections, inspire novel ideas, and create fruitful partnerships. We will investigate the above-mentioned thrusts on GAI→ED and ED→GAI along several topics related, but not limited, to:
- (GAI→ED) Sharing viewpoints, novel ideas, or field experiences about using generative AI in real-world educational settings.
- (GAI→ED) Exploring the capabilities of generative AI and large-language models in novel educational scenarios, e.g., personalized content generation and grading.
- (GAI→ED) Exploring novel human-machine collaborative systems where generative models play different roles, e.g., as digital tutors, assistants, or peers.
- (ED→GAI) Sharing viewpoints, unique challenges, or field experiences about concerns among educators and policymakers in using generative AI.
- (ED→GAI) Developing novel prompting and fine-tuning techniques to safeguard the outputs of generative AI and large-language models against biases and incorrect information.
- (ED→GAI) Developing novel safeguarding techniques to validate the authenticity of content, e.g., to determine whether an assignment was written by students or generated by models.
Schedule
You can access the latest schedule on the NeurIPS website.
Invited Speakers and Panelists
The workshop will have 6 talks from the invited speakers listed below; additionally, we will have a panel discussion with the speakers. Video recordings of talks and the panel will be publicly available on the NeurIPS website.
- Kristen DiCerbo. Khan Academy (Mountain View, CA, USA).
- Elena Glassman. Harvard University (Cambridge, MA, USA).
- Hieke Keuning. Utrecht University (Utrecht, Netherlands).
- Tobias Kohn. Karlsruhe Institute of Technology (Karlsruhe, Germany).
- Chris Piech. Stanford University (Stanford, CA, USA).
- Lu Wang. University of Michigan (Ann Arbor, MI, USA).
Papers
The workshop will have over 30 papers that authors will present in two poster sessions. You can find more details in the schedule on the NeurIPS website.
-
Paper 3: [Paper] [Poster] [Video] AuthentiGPT: Detecting Machine-Generated Text via Black-Box Language Models Denoising.
-
Paper 4: [Paper] [Poster] [Video] [Slide] Beyond Hallucination: Building a Reliable Question Answering & Explanation System with GPTs.
-
Paper 5: [Paper] [Poster] [Video] [Slide] Benchmarking Educational Program Repair.
-
Paper 7: [Paper] [Poster] [Video] [Slide] Neural Task Synthesis for Visual Programming.
-
Paper 8: [Paper] [Poster] [Video] Generative Agent for Teacher Training: Designing Educational Problem-Solving Simulations with Large Language Model-based Agents for Pre-Service Teachers.
-
Paper 9: [Paper] [Poster] [Video] [Slide] Angel: A New Generation Tool for Learning Material based Questions and Answers.
-
Paper 10: [Paper] [Poster] [Video] Automated Distractor and Feedback Generation for Math Multiple-choice Questions via In-context Learning.
-
Paper 11: [Paper] [Poster] [Video] EHRTutor: Enhancing Patient Understanding of Discharge Instructions.
-
Paper 12: [Paper] [Poster] [Video] Personalization and Contextualization of Large Language Models For Improving Early Forecasting of Student Performance.
-
Paper 14: [Paper] [Poster] [Video] [Slide] Code Soliloquies for Accurate Calculations in Large Language Models.
-
Paper 15: [Paper] [Poster] [Video] [Slide] Efficient Classification of Student Help Requests in Programming Courses Using Large Language Models.
-
Paper 16: [Paper] [Poster] [Video] [Slide] Diffusion Models in Dermatological Education: Flexible High Quality Image Generation for VR-based Clinical Simulations.
-
Paper 17: [Paper] [Poster] [Video] [Slide] Towards AI-Assisted Multiple Choice Question Generation and Quality Evaluation at Scale: Aligning with Bloom's Taxonomy.
-
Paper 18: [Paper] [Poster] [Video] [Slide] Small Generative Language Models for Educational Question Generation.
-
Paper 19: [Paper] [Poster] [Video] [Slide] Exploring Student-ChatGPT Dialogue in EFL Writing Education.
-
Paper 21: [Paper] [Poster] [Video] [Slide] Mathify: Evaluating Large Language Models on Mathematical Problem Solving Tasks.
-
Paper 26: [Paper] [Poster] [Video] The Behavior of Large Language Models When Prompted to Generate Code Explanations.
-
Paper 27: [Paper] [Poster] [Video] [Slide] Enhancing Writing Skills of Chilean Adolescents:
Assisted Story Creation with LLMs.
-
Paper 30: [Paper] [Poster] [Video] [Slide] Transforming Healthcare Education: Harnessing Large Language Models for Frontline Health Worker Capacity Building using Retrieval-Augmented Generation.
-
Paper 31: [Paper] [Poster] [Video] [Slide] Field Experiences and Reflections on Using LLMs to Generate Comprehensive Lecture Metadata.
-
Paper 32: [Paper] [Poster] [Video] [Slide] Conversational Programming with LLM-Powered Interactive Support in an Introductory Computer Science Course.
-
Paper 33: [Paper] [Poster] [Video] [Slide] WordPlay: An Agent Framework for Language Learning Games
-
Paper 34: [Paper] [Poster] [Video] [Slide] Are LLMs Useful in the Poorest Schools? TheTeacher.AI in Sierra Leone.
-
Paper 35: [Paper] [Poster] [Video] [Slide] Generative AI in the Classroom: Can Students Remain Active Learners?
-
Paper 38: [Paper] [Poster] [Video] [Slide] Ruffle&Riley: Towards the Automated Induction of Conversational Tutoring Systems.
-
Paper 39: [Paper] [Poster] [Video] [Slide] An Automated Graphing System for Mathematical Pedagogy.
-
Paper 40: [Paper] [Poster] [Video] [Slide] Retrieval-augmented Generation to Improve Math Question-Answering: Trade-offs Between Groundedness and Human Preference.
-
Paper 41: [Paper] [Poster] [Video] AI-Augmented Advising: A Comparative Study of ChatGPT-4 and Advisor-based Major Recommendations.
-
Paper 43: [Paper] [Poster] [Video] [Slide] Large Language Model Augmented Exercise Retrieval for Personalized Language Learning.
-
Paper 44: [Paper] [Poster] [Video] [Slide] Evaluating ChatGPT-generated Textbook Questions using IRT.
-
Paper 46: [Paper] [Poster] [Video] [Slide] Improving the Coverage of GPT for Automated Feedback on High School Programming Assignments.
-
Paper 47: [Paper] [Poster] Detecting Educational Content in Online Videos by Combining Multimodal Cues.
-
Paper 48: [Paper] [Poster] [Video] [Slide] GAI-Enhanced Assignment Framework: A Case Study on Generative AI Powered History Education.
Organizers
- Paul Denny. University of Auckland (Auckland, New Zealand).
- Sumit Gulwani. Microsoft (Redmond, WA, USA).
- Neil T. Heffernan. Worcester Polytechnic Institute (Worcester, MA, USA).
- Tanja Käser. EPFL (Lausanne, Switzerland).
- Steven Moore. Carnegie Mellon University (Pittsburgh, PA, USA).
- Anna N. Rafferty. Carleton College (Northfield, MN, USA).
- Adish Singla. Max Planck Institute for Software Systems (Saarbrucken, Germany).
Reviewers
- Husni Almoubayyed (Carnegie Learning)
- David Ifeoluwa Adelani (University College London)
- Tulasi B (Christ University)
- Sami Baral (Worcester Polytechnic Institute)
- Conrad Borchers (Carnegie Mellon University)
- Nigel Bosch (University of Illinois Urbana-Champaign)
- José Cambronero (Microsoft)
- Tianying Chen (Carnegie Mellon University)
- Steven Dang (Lexia Learning)
- Paul Denny (University of Auckland)
- Jibril Frej (EPFL)
- Wenbin Gan (National Institute of Information and Communications Technology)
- Zhikai Gao (North Carolina State University)
- Ahana Ghos (Max Planck Institute for Software Systems)
- Alkis Gotovos (Max Planck Institute for Software Systems)
- Rinkaj Goyal (Guru Gobind Singh Indraprastha University)
- Feifei Han (Australian Catholic University)
- Yomna Mahmoud Ibrahim Hassan (University of Prince Edward Island)
- Muntasir Hoq (North Carolina State University)
- Takoua Jendoubi (Imperial College London)
- Parameswaran Kamalaruban (Alan Turing Institute)
- Hieke Keuning (Utrecht University)
- Issarapong Khuankrue (King Mongkut's Institute of Technology Ladkrabang)
- Tobias Kohn (Karlsruhe Institute of Technology)
- Viraj Kumar (Indian Institute of Science)
- Juho Leinonen (University of Auckland)
- Qinyi Liu (University of Bergen)
- Qianou Ma (Carnegie Mellon University)
- Farnam Mansouri (University of Waterloo)
- Rabia Maqsood (National University of Computer and Emerging Sciences)
- Paola Mejia-Domenzain (EPFL)
- Steven Moore (Carnegie Mellon University)
- Ilya Musabirov University of Toronto)
- Tanya Nazaretsky (EPFL)
- Andrew Olney (University of Memphis)
- Victor-Alexandru Pâdurean (Max Planck Institute for Software Systems)
- Eduard Pogorskiy (Durham University)
- Seth Poulsen (Utah State University)
- James Prather (Abilene Christian University)
- Ethan Prihar (EPFL)
- Anna N. Rafferty (Carleton College)
- Brent Neal Reeves (Abilene Christian University)
- Jaromir Savelka (Carnegie Mellon University)
- Nicy Scaria (Indian Institute of Science)
- Yang Shi (North Carolina State University)
- Adish Singla (Max Planck Institute for Software Systems)
- Vinitra Swamy (EPFL)
- Georgios Tzannetos (Max Planck Institute for Software Systems)
- Sowmya Vajjala (National Research Council Canada)
- Christabel Wayllace (Washington University Saint Louis)
- Daniel Weitekamp (Carnegie Mellon University)
- Tonghui Xu (University of Massachusetts Lowell)
Call for Papers
Details about papers submissions are provided below:
- We invite submissions of research papers reporting new results and position papers reporting new viewpoints or field experiences. As per NeurIPS guidelines, previously published work or work that is presented at the NeurIPS conference is not acceptable as a workshop submission.
- Submissions are limited to 6 pages of main content, including all figures and tables; additional pages containing references are allowed. If authors wish to put supplemental information in the paper (e.g., implementation details, proofs), they can use additional pages after the references to add appendices. Note that this supplemental information will not be used during the reviewing process, and reviewers should be able to judge your work based on the main content (up to 6 pages).
- All submissions must be in PDF format based on the NeurIPS 2023 LaTeX style file.
- The reviewing process is double-blind. All submissions should be anonymous.
- Accepted papers will be made publicly available as non-archival reports.
- We are using OpenReview to manage workshop submissions. All authors must have an OpenReview profile when submitting. The author list cannot be changed after the deadline.
- Submission portal: NeurIPS'23 Workshop GAIED.
- Paper submission deadline: 25 September 2023, 01:00 PM PDT (Pacific Daylight Time).
- Author notification: 18 October 2023, 01:00 PM PDT (Pacific Daylight Time).
GAIED workshop welcomes participation from individuals who do not have something they'd like to submit but are interested in the workshop topics. The workshop aims to facilitate new connections, inspire novel ideas, and create fruitful partnerships.