This paper considers the development of an AI-based provably-correct mathematical proof tutor. While Large Language Models (LLMs) allow seamless communication in natural language, they are error prone. Theorem provers such as Lean allow for provable-correctness, but these are hard for students to learn. We present a proof-of-concept system (LeanTutor) by combining the complementary strengths of LLMs and theorem provers. LeanTutor is composed of three modules: (i) an autoformalizer/proof-checker, (ii) a next-step generator, and (iii) a natural language feedback generator. To evaluate the system, we introduce PeanoBench, a dataset of 371 Peano Arithmetic proofs in human-written natural language and formal language, derived from the Natural Numbers Game.

Undergraduate research experiences have been shown to increase student retention rates in STEM pathways, with a notable impact on students from Historically Underrepresented Groups (HUGs). However, undergraduate research experiences are often inaccessible to students, particularly in high-demand research areas and at large institutions with low faculty-to-student ratios.

As the place where so many individuals make lifelong friends, learn how to navigate the world, and grow into themselves, compassion must start in the classroom. However, current American classrooms sometimes lack the very principles of equality and inclusion which underlie compassionate communities. One such case can be seen in the systemic lingual ostracization of students who are not native English speakers (English language learners). As our student design partners indicated, students’ and teachers’ mutual understanding despite differences in language, contributes to a feeling of “love and comfort in the classroom”. Inspired by our student design partners’ proposal of the Robo-Assistant, we decided to answer the Interaction Design and Children (IDC) 2023 research and design challenge with GlotBot! GlotBot is a mobile application to be used by teachers while delivering classroom instruction. GlotBot is designed for the secondary mathematics classroom, a prime setting in which students’ English language proficiencies are unfairly conflated with their mathematical abilities. GlotBot translanguages; it generates hybrid translations of teacher speech in real-time by translating non-technical terms into Spanish, but keeping technical terms (“coordinate plane”, “graph”) in English. The teacher can choose to repeat this hybrid translation out loud, to create more access points to the content for English language learners. The main objective of GlotBot is to assist teachers in delivering an equitable pedagogy to create an inclusive and compassionate classroom space that celebrates our nation’s lingual diversity.

Student-led study groups often play a key role in augmenting the classroom experience, both academically and inter-personally, through the formation of academic and personal communities. However, many students, especially underrepresented minority (URM) students, often report challenges in finding and maintaining study groups.