ai-tutoring
Filtering by topic ai-tutoring(3)Clear all filters
- PaperarXiv — AI in Education (cs.CY)23 Jul 2026
AI Assistants Overassist
Verona Teo, Raghav Jain, Tobias Gerstenberg, Max Kleiman-Weiner
Large language models are increasingly used as tutors but tend to intervene too early and too frequently, often providing complete solutions instead of targeted hints. A new benchmark, Int-Bench, evaluates LLM interventions across three domains and finds that LLMs prioritize short-term task success over fostering deep learning. This overassistance contrasts with human behavior and may hinder cognitive engagement and long-term learning.
Original abstract
Large language models (LLMs) are increasingly used as tutors and thought partners, helping users reason through problems. While guidance from AI assistants can scaffold thinking and foster learning, such benefits depend on how they help--for instance, intervening too early or too frequently may hinder true learning and cognitive engagement. Yet how AI systems navigate intervention decisions during problem-solving remains poorly understood. Here, we introduce Int-Bench, a simulation-based benchmark for evaluating LLM interventions during learning. Int-Bench simulates a "student" solving a problem while a "teacher" monitors the student's reasoning and decides whether, when, and how to intervene. Across three domains--code debugging, mathematics, and brain teasers--we evaluate LLM teachers on the frequency and timing of interventions, as well as their impact on both immediate task success and generalization to new problems. We also compare LLMs to humans, finding that LLMs intervene more frequently and earlier than humans. Moreover, in contrast to humans, they tend to provide complete solutions rather than targeted hints. These findings suggest that current LLM assistants often optimize for short-term success rather than supporting the reasoning processes needed for deeper learning and long-term success.
- PaperarXiv — AI in Education (cs.CY)17 Jul 2026
EduGuard: A Safe RAG-Based LLM Tutor for Programming Education
S M Asif Hossain, Ruksat Khan Shayoni, M. F. Mridha, Jungpil Shin
EduGuard presents a safe retrieval-augmented generation (RAG) framework for introductory programming tutoring, integrating query understanding, instructor-approved course retrieval, pedagogical strategy selection, rubric-aware generation, claim-level verification, and overreliance control. Evaluated on a custom benchmark and a real-world course-forum set, it achieves 90.1% correctness and 89.4% grounding with low hallucination and direct-answer leakage. A small pilot study shows improved learning outcomes and reduced passive dependence compared to a baseline GPT-4o-mini Tutor.
Original abstract
Generative AI (GenAI) is increasingly used by students for programming explanation, debugging, and assignment support. Yet unrestricted large language model (LLM) tutors can hallucinate, contradict course policy, reveal complete solutions, and foster passive dependence. This paper presents EduGuard, a safe retrieval-augmented generation (RAG) tutoring framework for introductory programming. EduGuard integrates query understanding, instructor-approved course retrieval, pedagogical strategy selection, rubric-aware generation, claim-level verification, and overreliance control. To make evaluation provenance explicit, we construct BILearn-CS, a 600-query instructor-authored, TA-validated benchmark spanning concept questions, debugging cases, misconceptions, assignment-support requests, code-mixed Bangla-English queries, and adversarial direct-answer prompts. Moving beyond a synthetic-only benchmark, we further evaluate on a 150-query public CS50-style course-forum set and run a small controlled pilot with 10 undergraduates using a counterbalanced pre-test/post-test design. Using Meta-Llama-3.1-8B-Instruct as the primary generator, hybrid FAISS/BM25 retrieval, and DeBERTa-v3-large-MNLI as an architecturally separate verifier, EduGuard is compared against strong baselines: GPT-4o-mini Tutor, Llama Socratic Tutor, LPITutor-style RAG, RAG with rubric prompting, and RAG with same-model self-checking. On BILearn-CS, EduGuard attains the best correctness (90.1%), grounding (89.4%), and rubric alignment (90.8%), with the lowest hallucination (4.9%) and direct-answer leakage (9.8%). In the pilot, it raises immediate post-test accuracy from 68.4% to 81.2% and cuts overreliance from 38.0% to 17.0% relative to GPT-4o-mini Tutor. These results suggest safe GenAI tutoring requires not only retrieval or strong prompting, but explicit pedagogical control, evidence verification, and deployment safeguards.
- PaperComputer Assisted Language Learning11 Jun 2026
From declarative knowledge to procedural fluency: a generative AI tutor for Arabic agreement rules
Djemai Mahmoud Boulaares
This paper proposes a generative AI tutor designed to help Arabic learners transition from declarative knowledge of agreement rules to procedural fluency in applying them.