Temporal/Casual QA

NExT-QA is a VideoQA benchmark targeting the explanation of video contents. It challenges QA models to reason about the causal and temporal actions and understand the rich object interactions in daily activities, e.g., "why is the boy crying?" and "How does the lady react after the boy fall backward?". It supports both multi-choice and generative open-ended QA tasks. The videos are untrimmed and the questions usually invoke local video contents for answers.

This dataset is aimed to study the existing reading comprehension models' capability to perform temporal reasoning, and see whether they are sensitive to the temporal description in the given question.

Click to add a brief description of the dataset (Markdown and LaTeX enabled). Provide: * a high-level explanation of the dataset characteristics * explain motivations and summary of its content * potential use cases of the dataset

Question answering over temporal knowledge graphs (TKGs) is crucial for understanding evolving facts and relationships, yet its development is hindered by limited datasets and difficulties in generating custom QA pairs. We propose a novel categorization framework based on timeline-context relationships, along with \textbf{TimelineKGQA}, a universal temporal QA generator applicable to any TKGs. The code is available at: \url{https://github.com/PascalSun/TimelineKGQA} as an open source Python package. We present a comprehensive temporal question categorization framework and a universal TKGQA dataset generator that addresses the limitations of existing datasets. Our framework introduces a systematic approach to classify question complexity through context facts, answer focus, and temporal operations, while identifying four key temporal capabilities (TCR, TPR, TSO, and TAO). Using our \textbf{TimelineKGQA} generator to create two benchmark datasets, we demonstrate that question difficulty aligns with our complexity categorization through empirical evaluation. This work provides a foundation for developing and evaluating more advanced TKGQA solutions, and enables widespread application in private domains.