GPU hardware implements a SIMT execution model, where small groups of threads, called subgroups (or warps in CUDA), execute synchronously. Languages expose this through high-performance subgroup-level APIs. However, providing precise subgroup semantics in languages is challenging, as compilers may transform the program, potentially disrupting source-level synchronous behavior even if the hardware is synchronous. As a result, no GPU programming language provides rigorous semantics for subgroup execution. In this work, we present SIMT-Step, a formal and flexible operational semantics for subgroup execution. At its core is a new semantic object, dynamic basic blocks, which enables precise specification of converged subgroup execution. SIMT-Step then provides flexibility for the execution of instructions, which can be collective, synchronous, or independent. We propose several candidate instantiations of SIMT-Step and design a suite of idiomatic tests to distinguish them, highlighting counter-intuitive behavior that arises under relaxed variants. We implement SIMT-Step in TLA+ and validate the behavior of the idiomatic tests, all of which can be verified in under one second. Finally, to investigate how closely SIMT-Step models real-world GPU behavior, we conduct a large fuzzing campaign, spanning nine GPUs and seven vendors. Our results show that, despite hesitancy to provide guarantees in official specifications, most GPUs exhibit strongly synchronous behavior. Combined, these contributions provide both a theoretical foundation and practical tools for reasoning about subgroup semantics in GPU programming languages.

Large language models (LLMs) have shown remarkable performances in generic question-answering (QA) but often suffer from domain gaps and outdated knowledge in smart manufacturing (SM). Retrieval-augmented generation (RAG) based on LLMs has emerged as a potential approach by incorporating an external knowledge base. However, conventional vector-based RAG delivers rapid responses but often returns contextually vague results, while knowledge graph (KG)-based methods offer structured relational reasoning at the expense of scalability and efficiency. To address these challenges, a hybrid KG-Vector RAG framework that systematically integrates structured KG metadata with unstructured vector retrieval is proposed. Firstly, a metadata-enriched KG was constructed from domain corpora by systematically extracting and indexing structured information to capture essential domain-specific relationships. Secondly, semantic alignment was achieved by injecting domain-specific constraints to refine and enhance the contextual relevance of the knowledge representations. Lastly, a layered hybrid retrieval strategy was employed that combined the explicit reasoning capabilities of the KG with the efficient search power of vector-based similarity methods, and the resulting outputs were integrated via prompt engineering to generate comprehensive, context-aware responses. Evaluated on design for additive manufacturing (DfAM) tasks, the proposed approach achieved 77.8% exact match accuracy and 76.5% context precision. This study establishes a new paradigm for industrial LLM systems, which demonstrates that hybrid symbolic-neural architectures can overcome the precision-scalability trade-off in mission-critical manufacturing applications. Experimental results indicated that integrating structured KG information with vector-based retrieval and prompt engineering can enhance retrieval accuracy, contextual relevance, and efficiency in LLM-based Q&A systems for SM.

Translating natural language inquiries into executable Cypher queries (text-to-Cypher) is a persistent bottleneck for non-technical teams relying on knowledge graphs (KGs) in fast-changing industrial settings. Rule and template converters need frequent updates as schemas evolve, while supervised and fine-tuned parsers require recurring training. This study proposes a schema-guided prompting approach, namely text-to-Cypher with semantic schema (T2CSS), to align large language models (LLMs) with domain knowledge for producing accurate Cypher. T2CSS distils a domain ontology into a lightweight semantic schema and uses adaptive filtering to inject the relevant subgraph and essential Cypher rules into the prompt for constraining generation and reducing schema-agnostic errors. This design keeps the prompt focused and within context length limits while providing the necessary domain grounding. Comparative experiments demonstrate that T2CSS with GPT-4 outperformed baseline models and achieved 86 % accuracy in producing correct Cypher queries. In practice, this study reduces retraining and maintenance effort, shortens turnaround times, and broadens KG access for non-experts. • A T2CSS prompting approach that guides LLMs with the domain schema is proposed. • A systematic semantic schema to cover multifaceted concepts is designed. • An information filtering mechanism to select the relevant information is proposed. • Results achieve 86 % accuracy in translating user inquiries to Cypher statements.

The presence of Metaverse has elicited escalating attention in the next-generation Internet, followed by a large number of computationally intensive tasks, such as augmented reality, virtual reality, artificial intelligence-generated content (AIGC) applications, etc. With the popularity of mobile communication technology, mobile metaverse is becoming increasingly widespread. The resources required for these applications are rapidly growing in parallel with increasing demands from mobile Metaverse users (MUs), putting pressure on Metaverse service providers (MSPs) with limited resources, especially in mobile computing scenarios. Inspired by the burgeoning communication and computing technologies, the mobile Metaverse market between mobile MUs and MSPs is developing vigorously. However, there still remain numerous challenges in this market, including hierarchical mobile Metaverse structure, temporal dependencies, as well as heterogeneous incentive. In this paper, we propose a bi-alternating resource provisioning and heterogeneous auction approach for mobile Metaverse, named Bi-Meta. At the high level, resource provisioning is formulated as a Lyapunov problem minimizing average delay, solved by a novel bi-level based generative adversarial network, i.e., BiGAN. At the low level, a price- guided double dutch auction (PG-DDA) mechanism is presented for heterogeneous resource matching, with the designed PG- DDA smart contract. Based on the realistic edge-cloud company's traces, the experimental results verify that our proposed scheme achieves optimal latency and social welfare.

As the footing stone of artificial intelligence (AI), ubiquitous computing resource is beginning to receive interest. With this trend, a new form of edge-cloud service market dedicated to collecting, trading, and scheduling computing resources is rising. The computing participants in the service market, as providers and demanders of computing resources, are becoming more diversified and open. As such, the intriguing economic phenomenon and the circulation mechanism have emerged. These bring inherent challenges, such as a volatile market, the ossification of pricing, isolation, and inefficiency. In this article, we propose a novel space-time-request trading framework for the edge-cloud service market, namelyCompCube. To ensure stability,CompCubeadopts the dual-circulation futures-spot trading method, including space-time dynamic pricing in the macro-cycle, request intention conversion, and resource matching in the micro-cycle. Among this, an incomplete information game model is designed to determine the long-term trading pricing in the macro-cycle. Then, to tackle the indicator isolation problem due to the inconsistency between the user's requests and the computing-power provider's (CPP’s) resources, we focus on minimizing the rental cost of computing resources while satisfying diverse service level agreements (SLA) of users. To address this problem, a spatiotemporal scale Lyapunov optimization and an alternating actor-critic algorithm, A2SC, are developed. Besides, in the micro-cycle, a discriminatory double auction helps to determine the computing resource matching results efficiently and impersonally. We evaluate theCompCubeof the A2SC algorithm with realistic datasets. Compared to other baselines, i.e., DYRECEIVE, Price Preferred, and Random, A2SC reduces the average rental cost by 30.45%, 5.74%, and 17.57%, respectively. Furthermore,CompCubecan improve SLA satisfaction, as well as promote resource efficiency and social welfare compared with the traditional methods.