SQuAD

Dataset Card for SQuAD Dataset Summary Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowdworkers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question m

Papers using SQuAD (10)

Same Meaning, Different Scores: Lexical and Syntactic Sensitivity in LLM Evaluation2026Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation2026GradMem: Learning to Write Context into Memory with Test-Time Gradient Descent2026Probing for Knowledge Attribution in Large Language Models2026HaluNet: Learning Hallucination Risk from Internal Signals in LLM Question Answering2025A Context-Aware Dual-Metric Framework for Confidence Estimation in Large Language Models2025Cleanse: Uncertainty Estimation Approach Using Clustering-based Semantic Consistency in LLMs2025Growing Transformers: Modular Composition and Layer-wise Expansion on a Frozen Substrate2025Drop Dropout on Single-Epoch Language Model Pretraining2025Efficient Data Selection at Scale via Influence Distillation2025