REVIEW 25 cited by
Large Language Models for Education: A Survey and Outlook
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
The advent of Large Language Models (LLMs) has brought in a new era of possibilities in the realm of education. This survey paper summarizes the various technologies of LLMs in educational settings from multifaceted perspectives, encompassing student and teacher assistance, adaptive learning, and commercial tools. We systematically review the technological advancements in each perspective, organize related datasets and benchmarks, and identify the risks and challenges associated with deploying LLMs in education. Furthermore, we outline future research opportunities, highlighting the potential promising directions. Our survey aims to provide a comprehensive technological picture for educators, researchers, and policymakers to harness the power of LLMs to revolutionize educational practices and foster a more effective personalized learning environment.
Forward citations
Cited by 25 Pith papers
-
EduZone: A Framework for Evaluating LLM Safety for K-12 Students and Teachers
EduZone is a new evaluation framework and 5.2K-prompt dataset showing that LLMs are substantially more vulnerable to education-specific risks and adaptive multi-turn attacks than to conventional safety risks.
-
Beyond Semantic Equivalence: Logical Graphs for LLM Uncertainty Quantification
LGU models implication and incompatibility among LLM answers and reports consistent AUROC/AUARC gains over semantic entropy on QA benchmarks.
-
From Misunderstandings to Learning Opportunities: Leveraging Generative AI in Discussion Forums to Support Student Learning
M2M applies LLMs with retrieval-augmented generation to student forum posts to identify class-level misunderstandings and generate targeted learning resources, evaluated qualitatively with five instructors.
-
Hide and Seek with LLMs: An Adversarial Game for Sneaky Error Generation and Self-Improving Diagnosis
An adversarial training framework pairs a sneaky error generator with an error-diagnosing model, improving stealth and diagnosis on math problems.
-
Simplifications are Absolutists: How Simplified Language Reduces Word Sense Awareness in LLM-Generated Definitions
Simplified prompting (Simple and ELI5) sharply reduces the number of word senses LLMs provide for homonyms, and DPO fine-tuning of Llama 3.1 8B restores much of that completeness.
-
Insights in Adaptation: Examining Self-reflection Strategies of Job Seekers with Visual Impairments in India
Blind and visually impaired job seekers in India reflect on their job search mainly through social feedback, but such feedback is scarce, generic, or biased, and mainstream interview tools are largely inaccessible and...
-
COGENT: A Curriculum-oriented Framework for Generating Grade-appropriate Educational Content
Adding curriculum decomposition, readability constraints, and wonder-based topics to LLM prompts yields science reading passages with higher curriculum alignment and human-comparable comprehensibility.
-
From EduVisBench to EduVisAgent: A Benchmark and Multi-Agent Framework for Reasoning-Driven Pedagogical Visualization
EduVisAgent, a five-agent framework, outperforms all baseline AI models at generating pedagogically effective interactive visualizations for STEM problems, according to the new EduVisBench benchmark and its GPT-4o-bas...
-
CoTSRF: Utilize Chain of Thought as Stealthy and Robust Fingerprint of Large Language Models
CoTSRF fingerprints a source LLM by training a contrastive encoder on chain-of-thought responses, then flags suspect APIs whose reasoning-style feature distances are too close to the source's distribution.
-
Fine-Tuned LLMs are "Time Capsules" for Tracking Societal Bias Through Books
Fine-tuned LLMs reproduce era-specific stereotypes from bestselling fiction, but the lack of a direct book-content baseline makes the exact trends uncertain.
-
How Jailbreak Attacks Inform Safety Alignment: A Defender-Centric, Shapley-Based Evaluation of Jailbreak Contributions
Attacks that break LLMs best are not the ones that improve safety most; a Shapley- and greedy-based framework that selects attack subsets by downstream defender utility outperforms attacker-centric and attribution-onl...
-
Investigating Student Interaction Patterns with Large Language Model-Powered Course Assistants in Computer Science Courses
A deployed LLM course assistant served 589 students across three CS courses; logs show heavy evening use and homework questions, while only about 11% of responses included AI follow-ups that students mostly ignored.
-
ZPD-SCA: Unveiling the Blind Spots of LLMs in Assessing Students' Cognitive Abilities
ZPD-SCA, an expert-annotated Chinese reading benchmark, shows LLMs judge reading difficulty for student age groups poorly in zero-shot settings and improve, but remain biased, with in-context examples.
-
AI-Powered Math Tutoring: Platform for Personalized and Adaptive Education
A multi-agent math tutoring platform integrating Socratic prompting, GraphRAG retrieval, dual memory, and DAG course planning outperforms a baseline prompt on a simulated tutoring benchmark.
-
Can LLMs Reliably Simulate Real Students' Abilities in Mathematics and Reading Comprehension?
No LLM-prompt pair among 11 models and 4 prompts aligns with average NAEP student performance across math and reading in grades 4, 8, and 12.
-
FEAT: A Preference Feedback Dataset through a Cost-Effective Auto-Generation and Labeling Framework for English AI Tutoring
A cost-effective dataset framework for English AI tutoring finds that mixing 5-10% human-annotated feedback with LLM-generated feedback outperforms using only human-annotated feedback.
-
Benchmarking the Pedagogical Knowledge of Large Language Models
The authors release an open benchmark of 1,143 pedagogical knowledge questions from Chilean teacher exams and report accuracy, cost, and size trade-offs for 97 large language models.
-
"Check My Work?": Measuring Sycophancy in a Simulated Educational Context
Across five OpenAI models, mentioning a correct answer in a query boosts LLM accuracy by up to 15 points, while mentioning an incorrect answer lowers it by a similar amount.
-
Intent Matters: Enhancing AI Tutoring with Fine-Grained Pedagogical Intent Annotation
Fine-tuning a math tutor model on 11 fine-grained pedagogical intents instead of 4 broad ones gave better automatic scores and a modest human preference in a small evaluation.
-
Uni-Retrieval: A Multi-Style Retrieval Framework for STEM's Education
Uni-Retrieval uses a learnable Prompt Bank selected by prototype similarity to improve multi-style STEM image retrieval, evaluated on a new 24,000-query educational dataset.
-
VTutor: An Open-Source SDK for Generative AI-Powered Animated Pedagogical Agents with Multi-Media Output
VTutor is an open-source SDK that wires together LLMs, TTS, and uLipSync to produce animated pedagogical agents that speak with synchronized lip movements.
-
Beyond Grading Accuracy: Exploring Alignment of TAs and LLMs
Six open-source LLMs reach up to 88.56% per-criterion accuracy and Pearson r≈0.80 versus TA grades on 92 UML class diagrams, supporting mixed-initiative grading.
-
Using Large Language Models to Suggest Informative Prior Distributions in Bayesian Statistics
LLMs suggested directionally correct but poorly calibrated Bayesian priors, with Claude's weak priors ranking best on KL divergence from the data.
-
Temporalizing Confidence: Evaluation of Chain-of-Thought Reasoning with Signal Temporal Logic
Stepwise CoT confidence is reshaped and scored with signal temporal logic robustness to produce better calibrated confidence estimates on Gaokao math questions.
-
Towards Trustworthy Retrieval Augmented Generation for Large Language Models: A Survey
A survey organizes current research on trustworthy RAG into six pillars, reliability, privacy, safety, fairness, explainability, and accountability, and maps methods, metrics, and open problems for each.
Discussion (0). Continue with ORCID to comment.