In real-world business environments, data is stored in a variety of sources, including structured relational databases, semi-structured databases, and unstructured files. The ability to extract reasonable insights across these diverse sources is integral to data-driven decision-making. Existing benchmarks, however, are limited in assessing agents’ capabilities across these diverse data types. To address this gap, we introduce UniDataBench, a multi-source benchmark designed to evaluate the performance of data analytics agents in handling diverse data sources. Specifically, UniDataBench is constructed based on real-life industry analysis reports, employing a pipeline to synthesize data that aligns with authentic analytical trends. It encompasses diverse datasets spanning relational databases, CSV files, and NoSQL stores to reflect real-world business settings, and provides a unified framework for evaluating how effectively agents can explore multiple data formats, extract insights, and generate meaningful summaries and recommendations. Based on UniDataBench, we propose a novel LLM-based agent named ReActInsight, an autonomous agent that performs end-to-end analysis over diverse data sources by automatically discovering cross-source linkages, decomposing goals, and generating robust, self-correcting code to extract actionable insights. Our benchmark and agent together provide a framework for facilitating the development of data analytics agents in real-world applications.

Reinforcement learning (RL) has been widely adopted to enhance the performance of large language models (LLMs) on Text-to-SQL tasks. However, existing methods often rely on execution-based or LLM-based Bradley–Terry reward models. The former suffers from high execution latency caused by repeated database calls, whereas the latter imposes substantial GPU memory overhead, both of which significantly hinder the efficiency and scalability of RL pipelines. To this end, we propose a novel reward model framework for RL-based Text-to-SQL named Graph-Reward-SQL, which employs the GMNScore outcome reward model. We leverage SQL graph representations to provide accurate reward signals while significantly reducing time cost and GPU memory usage. Building on this foundation, we further introduce StepRTM, a stepwise reward model that provides intermediate supervision over Common Table Expression (CTE) subqueries. This encourages both functional correctness and readability of SQL. Extensive comparative and ablation experiments on standard benchmarks, including Spider and BIRD, demonstrate that our method consistently outperforms existing reward models.

By examination of the high-frequency nouns, verbs, and keywords, the present study probes into the similarities and differences of corporate images represented in Corporate Social Responsibility (CSR) reports of China Mobile and Vodafone. The results suggest that: 1) both China Mobile and Vodafone prefer using some positive words, like improve, support and service to shape a positive, approachable and easy-going corporate image, and an image of prioritizing the environmental sustainability and the well-being of people; 2) CSR reports of China Mobile contain the keywords poverty and alleviation, which means China Mobile is pragmatic, collaborative and active to assume the responsibility for social events; 3) CSR reports of Vodafone contain keywords like privacy, women and global as well as some other countries, which shows Vodafone is enterprising, globalized and attentive to the development of women; 4) these differences might be related to the ideology and social culture of Chinese and British companies. This study may contribute to understanding the function of CSR report and offer helpful implications for broadening the research of corporate image.