Natural Language to SQL (NL2SQL) enables natural language access to structured data, but LLM-based methods can be inefficient for real-time use and repetitive query patterns. We present AutoRuleSQL, a hybrid system that combines template-based fast paths with LLM fallback and offline bootstrapping. Empirical results show that it reduces latency by over 12.6% and improves execution accuracy by up to 4.0%, when combined with existing NL2SQL methods.

This paper presents a structured decoding approach to support reliable and efficient container orchestration using large language models (LLMs) in conjunction with the Model Context Protocol (MCP), a standard interface for LLMs to interact with Docker and Kubernetes. We address key challenges in using LLMs for container orchestration: high token overhead from outputs and the risk of generating invalid or unsafe commands. Empirical results demonstrate up to a 76.2% latency reduction.

Integrating Large Language Models (LLMs) with external tools and APIs is essential for fields such as information retrieval and knowledge management. While LLMs have made significant strides, their effective integration with external APIs-essential for real-world applications-remains challenging. This paper introduces RESTful-Llama, a novel method designed to empower open-source LLMs to accurately convert natural language instructions into well-formed RESTful API calls. Moreover, RESTful-Llama utilizes DOC-Prompt, a newly proposed technique for generating fine-tuning datasets from publicly available API documentation. Initial experiments demonstrate that RESTful-Llama significantly enhances the accuracy of generated REST API requests.

Bloom filter (BF) encodings are a proven method for comparing the similarity of records from multiple databases while maintaining privacy. This process, known as privacy-preserving record linkage (PPRL), is computationally expensive, especially with large datasets. To address this challenge, we have observed that BF encodings often exhibit sparsely filled patterns. Leveraging this insight, we introduce SparseBF, a scalable data structure that is space-optimized and maintains fast computation speed for PPRL. Compared to typical BF encodings, SparseBF brings in three improvements. First, SparseBF employs a hybrid storage scheme that selects the optimal storage component for BF encodings. Second, SparseBF utilizes its adaptive compressed sparse row storage to achieve lossless space compression, both when BFs are sparsely occupied and when they are densely populated. Third, SparseBF supports SIMD vector instructions to optimize the record linkage speed. Experiments show that, in sparsely filled scenarios, SparseBF outperforms the existing solution by up to 2.1x record linkage computation speed, while delivering up to 70.5% savings in storage space.

Large language models (LLMs) have transformed automated code generation. However, their high computational demands often lead to server overload and increased latency in SaaS deployments. To address this, we present SpeCoder, a framework that accelerates server-side code generation using speculative sampling (SpS) and supervised fine-tuning (SFT). SpS allows lower latency in the code generation, whereas SFT enables more personalized code generation tailored to the user's needs.

Large language models (LLMs) have shown immense potential for applications in information retrieval and knowledge management, but their computational and memory demands pose challenges for resource-constrained devices. In response, this work introduces an FPGA-based accelerator designed to improve LLM inference performance on embedded devices. We leverage quantization techniques, asynchronous computation, and a fully-pipelined accelerator to enhance efficiency. Our empirical evaluations, conducted using the TinyLlama 1.1B model on a Xilinx ZCU102 platform, demonstrate a 14.3-15.8x speedup and a 6.1x energy efficiency improvement over running exclusively on the ZCU102 processing system (PS).