In 2015, Lemos, Daniel, and Benatallah introduced a comprehensive taxonomy of Web service composition approaches, techniques, and tools. Given that ten years have passed and the recent emergence of artificial intelligence techniques and Large Language Models (LLMs) for service composition and systems engineering, there is a need to update such a classification. To this end, we perform a Systematic Literature Review (SLR) on natural language-based automated service composition approaches and filter for approaches not covered within the existing taxonomy. Specifically, we study 234 papers published from 2014 to May 2026 and identify 20 novel approaches that the original taxonomy inadequately covers, given the technological shifts. The result of the present study is a revised taxonomy integrating eight additional subcategories. The revised taxonomy provides a structured framework, making it suitable for future work on automated service composition with LLMs.
Unstructured financial texts provide event-related signals for event-aware forecasting in intelligent financial services, but transforming heterogeneous texts into structured predictive representations remains challenging. Existing studies often insufficiently distinguish event influence scopes, inadequately model event interactions, and rely on shallow fusion between textual events and numerical market signals, thereby limiting their use in service-oriented financial analytics. To address these limitations, we propose an LLM-based event-driven service framework for intelligent financial forecasting. The framework transforms financial news and corporate announcements into structured event representations and integrates them with numerical market features. A schema-guided extraction pipeline identifies predefined financial event types and structured arguments, organizes events into local and global categories, and encodes them using template-based descriptions. A multi-scale convolutional sparse attention module captures event dependencies and cross-level interactions, whereas adaptive gated fusion integrates event and market representations. Stock movement forecasting is used as a representative downstream service task. Experiments on CSI 300 constituent stocks from 2022 to 2024 show that the proposed method outperforms representative baselines, achieving an ACC of 56.06
Large language models now serve enterprise systems, where they plan work, execute handoffs, review artifacts, and improve prior outputs rather than only answer isolated chat requests. Most safety benchmarks test single prompts or jailbreak dialogues and miss how unsafe intent can flow through workflow history, inherited subtasks, and intermediate artifacts. We present JailFlowEval, a workflow-safety benchmark that turns harmful intents into stage-specific records for four common workflow contexts: plan, exec, review, and improve. Its two-phase pipeline builds workflow state and rewrites refusal-triggering text while keeping the unsafe goal; the released high-quality subset contains 744 workflow evaluation records. Tests on 10 models show that workflow-aware subtasks reveal unsafe continuation risks missed by prompt-level evaluation. The main risk context varies by model and metric. This setup maps risk by stage and helps place safeguards where unsafe goals persist as work moves through later service steps after the initial user request is processed.
Real-world time series data is often highly non-stationary, typically exhibiting low-frequency dominated long-term trends and periodic structures alongside high-frequency short-term fluctuations and noise. These distinct frequency components are highly coupled within the time domain. Existing time-domain decomposition methods typically use fixed-size smoothing windows for trend extraction. This approach often blurs information or attenuates local dynamic features, making precise frequency isolation difficult. Furthermore, it is challenging for a single model to simultaneously balance long-range dependency modeling and local detail modeling, thereby limiting the accuracy of long-term forecasting. To address these issues, this paper proposes a dual-branch time series forecasting approach based on frequency decomposition. First, Reversible Instance Normalization (RevIN) is applied to mitigate the non-stationary statistical distribution shifts in the data. Subsequently, Fast Fourier Transform (FFT) is utilized to decompose the sequence in the frequency domain before transforming it back into the time domain. During the forecasting phase, a dual-branch architecture is designed: a Low-frequency Patch-based Forecasting Block (PFB) processes the low-frequency components, while a High-frequency Convolutional Forecasting Block (CFB) handles the high-frequency components. This design not only captures long-term trends and periodic structures via the patch mechanism and self-attention, but also fully leverages dilated causal convolutions to characterize local abrupt changes and short-term volatility. Finally, the prediction results from both branches are fused and denormalized to yield the final output sequence. Experimental results on seven benchmark datasets demonstrate that the proposed method achieves significant performance improvements in long-term time series forecasting tasks.
Environmental, Social, and Governance (ESG) considerations have become increasingly important in modern financial management; however, many existing ESG-oriented analytical models emphasize predictive accuracy while offering limited interpretability and weak support for actionable financial decision-making. An ESG-aware intelligent financial management framework operating as a sequential pipeline is proposed in this study. First, TabNet employs a sparse attention mechanism for instance-wise feature selection of influential ESG and financial indicators, ensuring interpretability and transparent identification of key sustainability drivers. The selected features are then passed to a deep neural network–based regression model to capture complex nonlinear relationships between ESG attributes and financial performance outcomes and generate accurate predictions. Finally, a constrained decision optimisation layer transforms the predicted outputs into actionable capital allocation and investment prioritisation strategies under sustainability and risk constraints. Experimental evaluation conducted on a comprehensive ESG and financial performance dataset demonstrates that the proposed framework achieves improved predictive accuracy compared with conventional regression and machine learning baselines, with a root mean squared error of 4.8617. The results further highlight the varying influence of environmental, social, and governance factors on financial outcomes and confirm the interpretability benefits of sparse attention mechanisms. Overall, the proposed framework provides an effective and transparent decision-support tool for ESG-driven financial management, bridging the gap between predictive modelling and practical, sustainability-oriented financial decision optimisation.
