The recent advance of robotic has led to an increasing demand for control software that governs their complex operations and adaptability across diverse tasks. Componentization is an effective way for the development of control software, while it faces two main challenges. Firstly, it is not clear how to develop unified components for control software, which impedes the assembly of components from different environments and programming languages. Meanwhile, the assembly of software components significantly relies on the domain knowledge, which is inextricably tied to task-specific demands. In order to address the aforementioned challenges, this study introduces a large language model (LLM) enhanced componentized software development framework for the agile development of the control software for robotic system. Firstly, a unified component development approach is proposed to develop the components that can be assembled seamlessly. Subsequently, a LLM is fine-tuned to generate the workflow of the task, which allows the efficient assembly the unified components. Thirdly, the software components are assembled as control software. A robotic sorting task and grinding task are implemented to validate the effectiveness of the proposed framework. The results substantiate that the proposed framework can flexibly and accurately generate control software for various production tasks.

To address the difficulty of recognizing tail classes in Web service classification caused by long-tailed data distributions, a reinforcement learning-enhanced framework is proposed, which integrates multi-round data augmentation with adaptive loss optimization. A large language model (LLM) is employed as the core semantic generator, and a reinforcement learning agent adaptively adjusts class-specific augmentation ratios and filtering thresholds at each iteration based on the observed environment state, driving a closed-loop multi-round process of “generation-evaluation-filtering-feedback.” In parallel, a Chain-of-Thought-based reasoning mechanism is introduced to evaluate generated samples from multiple dimensions－including novelty, semantic consistency, and reasoning quality－thereby filtering out template-like and semantically drifting instances and progressively improving the training data distribution. During classifier training, class weights are dynamically computed from the frequency statistics of the augmented dataset at each iteration. A Top-k Near-Miss Focal Loss is further designed to jointly emphasize long-tailed classes and near-miss boundary samples, penalizing ambiguous semantic regions and enabling adaptive loss optimization tailored to long-tailed and hard examples. Experiments conducted on a real-world long-tailed Web service dataset and the PMTD (Productive Math Tutoring Dialogue) instructional dialogue dataset demonstrate that the proposed method outperforms mainstream baselines such as NCAL (Neural-Collapse-Advanced Personalized Learning), RGPT, SRaSLR (Social Relation Aware Service Label Recommendation Model) and LLMEmbed across multiple evaluation metrics. In particular, substantial improvements are observed for tail-class recognition: on several lightweight models, Macro- $F_1$ improves by up to 5 percentage points on the Web service dataset, and Weighted- $F_1$ increases by approximately 2-3 percentage points. These results verify the effectiveness of the proposed approach in mitigating the bias introduced by long-tailed distributions and provide a practical solution for intelligent recognition and classification of semantically sparse services in open environments.

In recent years, with the increasing complexity of software, traditional manual and partially automated testing methods have found it difficult to meet the efficiency requirements of system testing work, especially when dealing with rapidly changing and highly customized industrial software systems, where testing coverage and efficiency are often insufficient. Therefore, in this research, a component-based industrial software system test case generation method is proposed based on a large language model, aiming to improve the flexibility and efficiency of testing work. This method uses a large language model to convert user descriptions into test cases, combined with a step-by-step generation strategy, covering three stages: test case generation, correction, and test code generation. Firstly, the system starts from user input and generates preliminary test case data based on the knowledge base. Subsequently, the Petri net model is used to modify the generated test case paths to ensure the accuracy of the testing process. Finally, by modularizing the generation of core business code, the deployment of test cases can be accelerated. The experimental results show that this method demonstrates good automation capabilities in component-based industrial software system testing, significantly improving testing efficiency and coverage.