1 January 2025 in Intelligent Computing
This paper presents a comprehensive survey of the current status and opportunities for large language models (LLMs) in task planning, a sophisticated process of reasoning and decision-making that organizes a sequence of actions to accomplish predefined goals. Task planning centers on identifying suitable solutions for a task, with a specific emphasis on ensuring its successful completion. Although task planning plays a crucial role in enabling systems to function effectively in dynamic and complex environments, there is a lack of systematic reviews on this topic. We explore the theories, methodologies, and applications related to task planning with LLMs, highlighting the burgeoning development in this field and the interdisciplinary approaches that enhance their ability to complete tasks. This survey aims to systematize and clarify the fragmented literature, provide a systematic review that underscores the importance of task planning as a critical capability, and offer insights into future research directions and potential improvements. We hope to help researchers gain a clear understanding of the field and spark greater interest in this highly impactful research direction. A continuously updated resource is available in our GitHub repository at https://github.com/ZhaiWenShuo/Survey-of-Task-Planning .
21 November 2025 in Scientific Data
We introduce ChronoQA, a benchmark dataset for Chinese question answering focused on evaluating temporal reasoning in Retrieval-Augmented Generation (RAG) systems. Built from over 300,000 news articles published between 2019 and 2024, ChronoQA contains 5,176 questions covering absolute, aggregate, and relative temporal types, with both explicit and implicit time expressions. The dataset features both single- and multi-document scenarios, reflecting real-world requirements for temporal alignment and logical consistency. By providing structured evaluation across a wide range of temporal tasks, ChronoQA offers a dynamic, reliable, and scalable resource for benchmarking RAG systems in evolving knowledge environments.