Test whether systems can find critical exceptions and negations in exam rules and answer with supporting evidence.

Connect professional exam answers to evidence in textbooks, syllabi, and standards to build more trustworthy, traceable open-book QA systems.

Turn virtual experiment descriptions into structured records of measurements, observations, and conditions.

Train and evaluate reliable open-book exam assistants with cross-document questions, synthesized answers, and traceable supporting evidence.

Turn DELF and DALF exam materials into traceable, source-grounded Q&A for reliable preparation support.

Help models clarify misconceptions with textbook-based explanations and strengthen specialized concept reasoning.

Make SAT answer choice decisions more retrievable and explainable with evidence that supports correct answers and refutes distractors.

Connect GRE passages with external web evidence and information-boundary analysis to help models explain reading context with verifiable support.

This dataset is used to identify students' operational behavior in an experimental teaching environment.

Connect SAT editing questions with applicable rules and answer evidence to build grounded tutoring and explanation systems.

Turn terminology, concept relationships, and disciplinary context from digital textbooks into training data for specialized knowledge understanding.

Evaluate fact verification and evidence reasoning with claims from online course captions paired with web evidence.

Evaluate whether models can clarify question meaning and provide well-grounded answers while maintaining context across question bank dialogues.

Evaluate whether RAG systems retrieve complete, applicable evidence for SAT math solutions with step-by-step annotated evidence chains.

Train models on Digital SAT calculator rules and built-in tool details often missing from general math Q&A.

Connects safety standards, laboratory scenarios, and supporting evidence to strengthen comprehensive safety judgment in virtual labs.

Connect calibration, readings, units, and significant digits to help models detect inconsistencies in virtual experiment results.

Train models to solve graduate entrance exam mathematics problems with questions, worked solutions, and structured reasoning steps.

Train models to explain GRE math concepts, rule conditions, and tricky boundary cases with greater precision.

Pairing exam syllabus pages with evidence-grounded candidate Q&A, this dataset helps models answer questions about tested content and exam rules.