Lianshan Sun

Many traditional ceramic techniques are inscribed on UNESCO’s Intangible Cultural Heritage lists; yet, expert scarcity, long training cycles, and stylistic homogenization impede intergenerational transmission and innovation. Although large language models offer new opportunities, research tailored to ceramics remains limited. To address this gap, we first construct EvalCera, the first open-source domain large language model evaluation dataset for ceramic knowledge, image analysis, and generation, and conduct large-scale assessments of existing general large language models on ceramic tasks, revealing their limitations. We then release the first ceramics-focused training corpus for large language models and, using it, develop CeramicGPT, the first domain-specific large language model for ceramics. Finally, we built ChatCAS, a workflow multi-agent system built on CeramicGPT and GPT-4o. Experiments show that our model and agents achieve the best performance on EvalCera (A) and (B) text tasks as well as (C) image generation. The code is publicly available.

The digital health field's pursuit of massive, personalized healthcare continuously faces constraints from doctors' resources and capacity limitations. Recently, the emergence of large language models (LLMs), with their remarkable comprehension and processing abilities, has revolutionized digital health and enhanced massive, personalized healthcare. Although these LLMs have achieved significant advancements, they have also introduced inevitable hallucinations, which impact patient safety when used in massive applications. To address these challenges, this study proposes a digital hospital for a massive, personalized, reliable healthcare service named the Chat Chain-Brain-based Doctor (CHATCBD). In addition, this study transforms the LLM-based diagnostic process into a digital hospital architecture, designs a controllable AI agents framework, and develops a self-audit mechanism to enhance their reliability. The proposed CHATCBD uses blockchain technology to decentralize external regulation of the LLMs' personalized diagnoses. It introduces a blockchain-based personalized routing management mechanism to improve patient-centered decision-making and designs a blockchain-based audit framework based on a proposed mathematical model that ensures both the professionalism and honesty of audits, serving as a safety net for addressing LLM hallucinations. The results of extensive experiments conducted on 13 datasets from multiple perspectives demonstrate that the proposed CHATCBD system can significantly enhance the capabilities of LLMs in personalized healthcare. The code used in this study can be found at https://github.com/LDDLQ/ChatCBD.

The problem of data privacy protection in the information age deserves people’s attention. As a distributed machine learning technology, federated learning can effectively solve the problem of privacy security and data silos. Differential privacy(DP) technology is applied in federated learning(FL). By adding noise to raw data and model parameters, it can further enhance the degree of data privacy protection. Over the years, differential privacy technology based on federated learning framework has been developed, which is divided into central differential privacy federated learning(CDPFL) and local differential privacy federated learning(LDPFL). Although differential privacy may reduce the accuracy and convergence of federated learning models while protecting data privacy, researchers have proposed a variety of optimization methods to balance privacy protection and model performance. This paper comprehensively expounds the research status of differential privacy techniques based on the federated learning framework, first providing detailed introductions to federated learning and differential privacy technologies, and then summarizing the development status of two types of federated learning differential privacy(DPFL) techniques respectively; for CDPFL, the paper divides the discussion into first proposal of CDP and typical application examples, the impact of Gaussian mechanisms on model accuracy, optimization based on asynchronous differential privacy, and insights from other scholars; for LDPFL, the paper divides the discussion into first proposal of LDP and typical application examples, processing multidimensional data and improving model accuracy, existing methods and optimization for reducing communication costs, balancing privacy protection and data usability, LDPFL based on the Shuffle model, and insights from other scholars; following this, the paper addresses and summarizes the unique challenges introduced by incorporating differential privacy into federated learning and proposes solutions; finally, based on a summary of existing optimization techniques, the paper outlines future directions and specifically discusses three research ideas for enhancing the optimization effects of federated differential privacy: advanced optimization strategies combining Bayesian methods and the Alternating Direction Method of Multipliers (ADMM), integrating lattice homomorphic encryption techniques from cryptography to achieve more efficient differential privacy protection in federated learning, and exploring the application of zero-knowledge proof techniques in federated learning for privacy protection.

Electronic health records (EHRs) support digital healthcare, precision medicine, and predictive analytics. Traditionally, EHRs’ distribution in centralized databases limits sharing across organizations. Also, the poor integration and limited patient control of EHRs obstruct a comprehensive view of health conditions, to the detriment of both patients and doctors. Blockchain technology promises decentralized patient-centric EHR management. However, decentralized solutions still face challenges in efficiently tracking and sharing scattered and correlated medical records. Patients struggle to construct and maintain the lineage of their overall medical records, so these solutions fail to efficiently and securely share correlated EHRs in a lineage with trusted doctors. Doctors from various departments struggle to access the most relevant EHRs necessary for a precise understanding of patients’ medical histories. This study introduces a blockchain-based e-healthcare system with provenance awareness, collecting data provenance to document traceable patient histories, facilitate rapid authorizations for correlated EHRs in a lineage, and conduct post-authorization audits. Designing a DAG-like data structure, the system efficiently stores the EHR provenance on blockchain and optimizes provenance-specific search. The system introduces a dedicated EHR authorization layer to facilitate dynamic authorization propagation along with the EHR provenance path. Harnessing the principles of Nash equilibrium, the system establishes a rigorous, honesty-driven auditing mechanism. This is done to scrutinize doctors’ data access provenance and the intricate journey of patient health records. Our model combines data provenance with blockchain, ensuring prompt, reliable EHR access. We developed a prototype using Ethereum blockchain smart contracts, finding that our model’s feasibility and performance align with expectations.

Access control is widely used technology for securing sensitive resources of information systems, such as personal data managed by cloud-based data store and sensitive data stream collected by smart devices. Existing access control systems mainly adopt centralized architecture and static access control models, including Access Control List, Role-based Access Control and Attribute-based Access Control. However, these systems fail to meet the increasing requirements of behavior based dynamic access control or requirements of owner initiated autonomous access control without relying on trustworthy third parties and suffer inherent drawbacks of single point of failure or dishonesty. To this end, a novel blockchain-based and provenance enabled dynamic access control scheme called BPDAC is proposed. Specifically, it collects and stores data provenance on blockchain to enable behavior-based dynamic access control; in particular, the quick lookup table structure is designed to speed up access control evaluation based on provenance with increasing complexity. It also provides specifications for formulating access control policies based on provenance. It utilizes a set of smart contracts on blockchain to enable decentralized and reliable autonomous access control. A prototype system is implemented on the Hyperledger Fabric and experiments are conducted to show that the proposed scheme is practically feasible and scalable in terms of the performance metrics of throughput and latency.

Binary translation is an important technique for achieving cross-architecture software migration. However, mainstream dynamic binary translation frameworks, such as QEMU, often generate a large amount of redundant code, which degrades the efficiency of the target code. To this end, we propose a dynamic–static binary translation method based on branch prediction. It first identifies parts of translation blocks following static branch prediction techniques. Then it translates these translation blocks into less-redundant native code blocks by canonical static translation algorithms. Finally, it executes all code blocks that are translated either statically or dynamically by correctly maintaining and switching their running contexts. In order to correctly weave the two types of translation activities, the proposed method only translates the next translation block that is data-independent from the current one by the active variable analysis algorithm, and records and shares the intermediate states of the dynamic and static translation activities via a carefully designed data structure. In particular, a shadow register-based context recovery mechanism is proposed to correctly record the running context of static translation blocks, and to correctly recover the context for dynamically translating and running blocks that were not statically translated. We also designed an adaptive memory optimization mechanism to dynamically release the memory of the mispredicted translation blocks. We implemented a dynamic–static binary translation framework by extending QEMU, called BP-QEMU (QEMU with branch prediction). We evaluated the translation correctness of BP-QEMU using the testing programs for the ARM and PPC instruction sets from QEMU, and evaluated the performance of BP-QEMU using the CoreMark benchmark code. The experimental results show that BP-QEMU can translate the instructions from the ARM and PPC architectures correctly; moreover, the average execution efficiency of the CoreMark code on BP-QEMU improves by 13.3% compared to that of QEMU.

This paper presents an improved YOLOv3 network, named MSFF-YOLOv3, for precisely detecting variable surface defects of aluminum profiles in practice. First, we introduce a larger prediction scale to provide detailed information for small defect detection; second, we design an efficient attention-guided block to extract more features of defects with less overhead; third, we design a bottom-up pyramid and integrate it with the existing feature pyramid network to construct a twin-tower structure to improve the circulation and fusion of features of different layers. In addition, we employ the K-median algorithm for anchor clustering to speed up the network reasoning. Experimental results showed that the mean average precision of the proposed network MSFF-YOLOv3 is higher than all conventional networks for surface defect detection of aluminum profiles. Moreover, the number of frames processed per second for our proposed MSFF-YOLOv3 could meet real-time requirements.

Traditional flowchart recognition methods have difficulties in detecting newly added symbols and distinguishing targets from complex backgrounds like line texture. Existing deep-learning-based object detectors and line segment detectors are promising in recognizing and distinguishing targets from texture backgrounds. However, using two separate detectors will inevitably cause unnecessary training and inference costs. Moreover, the insufficient volume and diversity of currently available dataset limit the effectiveness of model training. To address these issues, this paper proposes an end-to-end multi-task network FR-DETR (Flowchart Recognition DEtection TRansformer) and a new dataset for precise and robust flowchart recognition. FR-DETR comprises a CNN backbone and a shared multi-scale Transformer structure to perform symbol detection and edge detection using shared feature maps and respective prediction heads in a coarse-to-fine refinement process. The coarse stage analyzes features with low resolution and suggests candidate regions that contain potential targets for the fine stage to produce accurate predictions using features with high resolution. Meanwhile, a new dataset is constructed to provide more symbol types and complex backgrounds for network training and evaluation. It contains more than 1000 machine-generated flowchart images, 25K+ symbol instances with nine categories, and 20K+ line segments. The experiments show that FR-DETR achieves an overall precision and recall of 94.0% and 93.1% on the proposed dataset, and 98.7% and 98.1% on the CLEF-IP dataset, respectively, which all outperform the prior methods.

In the Big Data era, data provenance has become an important concern for enhancing the trustworthiness of key data that are rapidly generated and shared across organizations. Prevailing solutions employ authoritative centers to efficiently manage and share massive data. They are not suitable for secure and trustworthy decentralized data provenance sharing due to the inevitable dishonesty or failure of trusted centers. With the advent of the blockchain technology, embedding data provenance in immutable blocks is believed to be a promising solution. However, a provenance file, usually a directed acyclic graph, cannot be embedded in blocks as a whole because its size may exceed the limit of a block, and may include various sensitive information that can be legally accessed by different users. To this end, this paper proposed the BSTProv, a blockchain-based system for secure and trustworthy decentralized data provenance sharing. It enables secure and trustworthy provenance sharing by partitioning a large provenance graph into multiple small subgraphs and embedding the encrypted subgraphs instead of raw subgraphs or their hash values into immutable blocks of a consortium blockchain; it enables decentralized and flexible authorization by allowing each peer to define appropriate permissions for selectively sharing some sets of subgraphs to specific requesters; and it enables efficient cross-domain provenance composition and tracing by maintaining a high-level dependency structure among provenance graphs from different domains in smart contracts, and by locally storing, decrypting, and composing subgraphs obtained from the blockchain. Finally, a prototype is implemented on top of an Ethereum-based consortium blockchain and experiment results show the advantages of our approach.