Multi-model databases and polystore systems are increasingly studied for managing multimodel data holistically.As their primary interface, multi-model query languages (MMQLs) often exhibit complex grammars, highlighting the need for effective Text-to-MMQL translation methods.Despite advances in natural language translation, no effective solutions for Text-to-MMQL exist.To address this gap, we formally define the Text-to-MMQL task and present the first Text-to-MMQL dataset involving three representative MMQLs.We propose an adaptive Text-to-MMQL framework that includes both a schema embedding module for capturing multi-model schema information and an MMQL representation strategy to generate concise intermediate query formats with error correction in generated queries.Experimental results show that the proposed framework achieves over a 9% accuracy improvement over our adapted baseline methods.

Procedural Language extensions of SQL (abbr. PL/SQL) enhance database programming by integrating procedural constructs with SQL's declarative syntax, thereby improving the reusability, modularity, and maintainability of SQL. Besides, PL/SQL in database systems presents significant challenges in real-world development, primarily due to the inherent complexity of programming. To reduce the development difficulty of PL/SQL, this paper studies the novel task of translating natural language (NL) to PL/SQL (i.e., NL-to-PL/SQL), aimed at simplifying PL/SQL development. Recent advancements in language models have shown promise in translating natural language questions into SQL queries (i.e., Text-to-SQL). However, the state-of-the-art Text-to-SQL methods focus only on single SQL queries, neglecting the procedural extensions of SQL, which limits their effectiveness for the NL-to-PL/SQL task. In this paper, we propose PLForge, a suite of pre-trained language models with parameter configurations of 3B, 7B, and 15B, tailored for NL-to-PL/SQL tasks. To enhance the PL/SQL generation capabilities of PLForge, we leverage a curated PL/SQL-centric data corpus and employ an incremental pre-training approach. Furthermore, to fully exploit the potential of PLForge, we propose a comprehensive prompt construction strategy tailored specifically for PL/SQL. Given the scarcity of NL-to-PL/SQL datasets, we develop a template-based method for generating NL-to-PL/SQL data. We conduct a series of experiments on PLForge and several baseline models. Based on execution match and exact match metrics that are designed specifically for the NL-to-PL/SQL task, the experimental results demonstrate that PLForge outperforms existing models in both in-context learning and supervised fine-tuning settings.

Distinguished from terrestrial edge intelligence, satellite edge intelligence has unique characteristics, including the rapid mobility of satellites, limitations in computing and energy resources, and differences in the artificial intelligence models deployed on user devices, satellites, and ground cloud servers. In this paper, we propose a Deep Reinforcement Learning (DRL)-based resource management scheme for task inference in Low Earth Orbit (LEO)-based satellite-ground collaborative networks. In our approach, the task of a user can be inferred by the user device itself, the edge server of the current satellite via user-to-satellite transmission, the edge server of a neighboring satellite via satellite-to-satellite transmission, or a ground cloud server via satellite-to-cloud transmission. Our scheme jointly optimizes task offloading, computing resource allocation, and communication resource allocation to minimize the total system cost, which encompasses trade-offs among the task inference delays for all tasks, the energy consumption of system, and the task inference accuracies for all tasks, while ensuring that the transmit power budgets of all satellites and the satellite coverage time constraints for each user are met. A DRL-based algorithm combining the Softmax Deep Double Deterministic Policy Gradients (SD3) algorithm and two numerical methods is designed to solve the optimization problem efficiently. We prove the convergence of our algorithm and demonstrate the superiority of our scheme by performing extensive simulations in 4 scenarios with 4 reference schemes.

Recent work has shown the existence of an implicit dimension structure within coevolution problems, which can uniquely determine the overall performance of a individual. In this paper, we present a reliable dimension identifying method. Based on this , we put forward a suitable archive scheme, which maintains only the most representative individuals in terms of problem dimensional information identified during coevolution, and achieves minimum size while guaranteeing monotonic progress. The experimental results on COMPARE-ON-ONE and COMPARE-ON-ALL demonstrate the viability of the algorithm