In the era of fintech services computing, the convergence of advanced technologies has become pivotal for driving innovation. The seamless integration of IIoT-enabled data streams with artificial intelligence (AI) systems presents unprecedented opportunities for revolutionizing financial risk assessment. Efficient AI-to-AI business platforms are crucial for sustainable and transparent fintech credit risk analysis. However, current practices rely on single AI models, leading to inefficiencies and regulatory challenges. This paper proposes a novel Blockchain Regulatory Framework to address these issues, particularly in lending to small and micro enterprises. The key novelty of our work, compared to existing state-of-the-art, is fourfold: 1) a multi-level guarantee strategy that combines blockchain with AI-AI models to enable distributed risk-sharing among guarantee agencies; 2) a dual-chain blockchain structure for secure and efficient storage of credit data and processes; 3) the integration of Intel SGX with Oblivious RAM (ORAM) to ensure unprecedented confidentiality, integrity, and availability in credit data sharing; and 4) the incorporation of game theory to mathematically validate the anti-collusion properties of the AI-AI decision mechanism. Our findings indicate that the proposed framework offers a comprehensive solution for achieving safe, efficient, and transparent credit assessments. This approach holds significant promise for developing effective financing solutions while ensuring data security, regulatory compliance, and robust credit evaluation methodologies.

Practical applications often require forecasting the future states of short time series (STS) using multiple related long time series (LTS) as auxiliary data, a process known as few-shot prediction. The primary challenge, given the limited data on STS, is effectively capturing the pattern similarities between STS and LTS. Current methods, despite notable advancements, primarily focus on trans- ferring pattern characteristics from LTS to STS without explicitly addressing their similarities at various levels. To overcome this lim- itation, we propose a novel few-shot time series forecasting model called DANet. Drawing on the Retrieval-Augmented Generation (RAG) framework in large language models, DANet retrieves long and short sequences from LTS that closely resemble STS, thereby enhancing prediction accuracy while simultaneously reducing un- certainty through this retrieval process. First, we define two metrics to quantify pattern similarities between STS and LTS, addressing the issue of different representations of the same pattern due to variations in sequence length. Second, we propose a dual-attention mechanism which embeds the two similarities metrics to extract and integrate long and short sequences from LTS across variable and temporal levels for generating predictions. Our experiments across six scenarios show that DANet significantly outperforms six state-of-the-art (SOTA) methods.

Driven by the applications in autonomous driving, robotics, and augmented reality, 3D object annotation is a critical task compared to 2D annotation, such as spatial complexity, occlusion, and viewpoint inconsistency. The existing methods relying on single models often struggle with these issues. In this paper, we introduce Tri-MARF, a novel framework that integrates tri-modal inputs (i.e., 2D multi-view images, text descriptions, and 3D point clouds) with multi-agent collaboration to enhance the 3D annotation process. Our Tri-MARF consists of three specialized agents: a vision-language model agent that generates multi-view descriptions, an information aggregation agent that selects optimal descriptions, and a gating agent that aligns text descriptions with 3D geometries for more refined captioning. Extensive experiments on the Objaverse-LVIS, Objaverse-XL, and ABO datasets demonstrate the superiority of our Tri-MARF, which achieves a CLIPScore of 88.7 (compared to 78.6–82.4 for other SOTA methods), retrieval accuracy of 45.2/43.8 (ViLT R@5), and an impressive throughput of 12,000 objects per hour on a single NVIDIA A100 GPU.