Text-to-SQL translates natural language questions into executable SQL queries, enabling intuitive database access for non-experts. While large language models achieve strong performance on Text-to-SQL with prompting, they still struggle with complex queries that involve deeply nested logic or multiple clauses. A widely used approach employs SQL skeletons--intermediate representations of query logic--to streamline generation, but existing methods are limited by their reliance on a single structural hypothesis and lack of progressive reasoning. To overcome these limitations, we propose LEAF-SQL, a novel framework that reframes skeleton prediction as a coarse-to-fine tree search process. LEAF-SQL enables systematic exploration of diverse structural hypotheses with adaptive refinement. Several key techniques are employed in LEAF-SQL: (1) a three-level skeleton hierarchy to guide the search, (2) a Skeleton Formulation Agent to generate diverse candidates, and (3) a Skeleton Evaluation Agent to efficiently prune the search space. This integrated design yields skeleton candidates that are both structurally diverse and granularity-adaptive, providing a stronger foundation for the SQL generation. Extensive experiments show that LEAF-SQL consistently improves the performance of various LLM backbones. On the official hidden test set of the challenging BIRD benchmark, our method achieves 71.6 execution accuracy, which outperforms leading search-based and skeleton-based methods, affirming its effectiveness for complex queries.

Network intrusion detection systems (NIDS) are essential for web security by identifying and dropping malicious traffic. Existing in-network NIDS leverage the Tbps-level packet processing capability of programmable switches to achieve high-speed flow classification. They translate complex trained machine learning models to decision trees (DTs), where DTs are deployed on programmable switches via single-DT or multiple-DT deployment. However, they face a fundamental trade-off: single-DT deployment suffers from low classification accuracy due to over-pruning of trees, while multiple-DT deployment suffers from high overhead due to deploying multiple tree replicas. In this paper, we propose Proteus, an in-network malicious traffic detection system that achieves both high classification accuracy and low overhead. Its key idea is to split the original DT into critical and normal sub-trees, where these sub-trees have different impacts on overall accuracy. More precisely, Proteus first splits a DT into one critical and several normal sub-trees for adapting to the accuracy requirement and switch resource budgets. Second, it minimizes coordination overhead between sub-trees while ensuring full flow coverage via mixed-integer linear programming. Third, it dynamically reallocates or migrates sub-trees to adapt to changing resources by monitoring both classification accuracy and switch resource changes. Testbed experiments with 12.8 Tbps programmable switches show that Proteus improves classification accuracy, reduces switch resource consumption, and reduces classification latency.

Leather is among the most widely traded textile materials worldwide, and detecting surface defects is crucial to ensuring its quality. Conventional manual inspection of leather surfaces is labor-intensive, time-consuming, and prone to human error. Advances in computer vision have facilitated the development of automated leather surface defect detection. However, vision-based methods remain sensitive to lighting conditions, particularly on high-gloss leather surfaces. In this paper, we introduce Le-Radio, an industrial IoT (IIoT) oriented RF-based system for automated leather defect detection. Le-Radio provides a gloss-robust, ubiquitous, and continuous inspection solution, which is crucial for integrating quality control into smart factory pipelines and advancing the digital transformation of the traditional leather industry. To extract informative representations of leather defects from wireless signals, we employ continuous radio snapshots and refined signal features to differentiate defect patterns. Specifically, we design an imaging algorithm that continuously visualizes defects on moving leather samples using a fixed radar. To enhance Le-Radio’s robustness in diverse environments, we train a transferable model that maintains consistent detection performance across different scenarios. We conduct extensive experiments in three rooms using 120 leather samples to evaluate Le-Radio. Experimental results demonstrate that Le-Radio accurately detects diverse defects in high-gloss leather. Beyond leather inspection, this work establishes Le-Radio as a practical IoT edge sensing node for material surface inspection in the IIoT. By integrating RF-based defect scanning capabilities and a robust radar imaging framework into a unified system, this work establishes RF sensing as a viable and novel paradigm for material surface inspection in the IIoT, particularly in scenarios where optical methods are fundamentally limited.

This paper proposes a reinforcement learning (RL) framework to optimize sequential experimental design for complex industrial systems. The framework addresses efficient identification of unknown parameters in complex industrial systems under budget constraints. A deep neural network approximates the action-value function, while Proximal Policy Optimization (PPO) with a clipped surrogate objective optimizes the policy, represented by a separate deep neural network. The policy maps the posterior over unknown parameters to an optimal experimental input. Evaluated on a one-dimensional thermal conduction problem, the method strategically positions a temperature sensor to estimate unknown parameters of an instantaneous heat source. Simulation results demonstrate that the policy effectively guides experiments towards informative measurements, concentrating the posterior around true parameter values. Compared to a genetic algorithm, the RL approach achieves superior performance, highlighting its potential for optimizing sequential experimental design in complex, resource-constrained scenarios.

Deploying data plane programs across the network is typically formulated as a mixed-integer programming task, leading to a long execution time. In response, existing studies carefully tailor heuristics for specific task properties such as objectives. However, they suffer from poor solution quality under dynamic task deployment since they overfit specific task properties. Recently, generative diffusion models have been widely adopted in network optimizations due to their strong adaptability and generalization. Accordingly, in this poster, we propose a diffusion model-based framework for data plane program deployment tasks. Our key idea is to leverage the reverse denoising process of diffusion models to react to dynamic task changes at runtime while maintaining high solution quality. Preliminary results on our testbed show that we reduce latency by 66.67% and resource overhead by 58.62% during dynamic deployment.

Large language models (LLMs) with prompting have achieved encouraging results on many natural language processing (NLP) tasks based on task-tailored promptings. Text-to-SQL is a critical task that generates SQL queries from natural language questions. However, prompting on LLMs haven’t show superior performance on Text-to-SQL task due to the absence of tailored promptings. In this work, we propose three promptings specifically designed for Text-to-SQL: SL-prompt, CC-prompt, and SL+CC prompt. SL-prompt is designed to guide LLMs to identify relevant tables; CC-prompt directs LLMs to generate SQL clause by clause; and SL+CC prompt is proposed to combine the strengths of these above promptings. The three prompting strategies makes three solutions for Text-to-SQL. Then, another prompting strategy, the RS-prompt is proposed to direct LLMs to select the best answer from the results of the solutions. We conducted extensive experiments, and experimental results show that our method achieved an execution accuracy of 86.2% and a test-suite accuracy of 76.9%, which is 1.1%, and 2.7% higher than the current state-of-the-art Text-to-SQL methods, respectively. The results confirmed that the proposed promptings enhanced the capabilities of LLMs on Text-to-SQL. Experimental results also show that the granularity of schema linking and the order of clause generation have great impact on the performance, which are considered little in previous research.