LiDAR-inertial odometry is widely used in robotics navigation, autonomous driving, and drone operation to provide precise, low-latency motion estimation. Filter-based methods are fast but suffer from significant cumulative errors. Graph optimization methods reduce cumulative errors through loop closure detection but are computationally expensive. In this work, we propose LIO-DPC, a framework that combines the benefits of the filter-based approach and graph-based approach. First, we propose a dynamic pose chain optimization method. It generates an initial pose chain using the fast filter. This is followed by applying computationally efficient local graph optimization to a set of local pose chains to generate refined relative poses, which are then used to update the motion estimation. Second, we propose a loop sparsification approach to select representative loops that are both temporally and spatially proximate, to reduce the computational complexity in graph optimization and minimize loop errors. Extensive experiments demonstrate that LIO-DPC achieves real-time performance and outperforms state-of-the-art methods in accuracy.

With the advances of deep learning, efficient model inference is crucial. Deep learning compilers optimize inference by decomposing models into subgraphs and searching schedules for them, whose evaluation relies on accurate cost models. Existing methods suffer from high transformation overheads or limited prediction accuracy caused by insufficient structural representation of subgraphs and schedules. To address these limitations, we propose CAST, a framework that predicts schedule performance based on Abstract Syntax Trees (ASTs). CAST proposes AST classification based on structural similarity and class-specific cost models. Experiments show CAST achieves significantly reduced prediction errors and up to$13 \times$higher efficiency than prior methods.

With the widespread application of deep neural networks, improving inference efficiency has become increasingly critical. To speed up the inference, deep learning compilers search for high-performance schedules for the DNN tensor programs. During the process, cost models have been extensively studied to evaluate the performance of the schedules, such that high-performance ones can be efficiently obtained. However, existing methods suffer from either high overhead or low accuracy of performance evaluation, both of which limit the efficiency of the final schedule. To address these issues, we propose FASP, a fast and accurate framework for schedule performance evaluation, based on Abstract Syntax Trees (ASTs). First, we propose a redundancy-aware ASTs reduction method to generate compact ASTs for more accurate feature extraction. Second, we propose a feature extraction method based on compact ASTs, which extracts features by accounting for computation nodes, loop nodes, and their structural relationships. Third, we propose a composition-similarity-driven ASTs classification method and a class-specific cost model architecture for more accurate performance evaluation. FASP overcomes the limitations of prior methods by significantly reducing evaluation errors. Experiments show its excellent performance in both single-model and cross-model evaluation, with errors ranging from 6 % to$\mathbf{1 3 \%}$. Moreover, FASP can obtain high-performance schedules with$13 \times$lower latency.

Smart contracts are required to be instantiated in the predeployed stage, which consumes computation resources from then on. It is a big waste in the blockchain whose nodes are composed of IoT devices, as those devices often have limited resources (such as limited power supplies or a limited number of processes to run). Meanwhile, IoT devices are heterogeneous and different smart contracts are required. If those smart contracts are instantiated previously, numerous meaningless addresses are required. In this paper, we propose to delay the instantiation of a smart contract when used and terminate it when not used, which is similar to the life cycle of a variable. Then, a new kind of variable (the wrapping variable) is used to hide details of the instantiation and the address. The smart contract is instantiated in the construction function of the wrapping variable, or even it is delayed to the time when there are requests for it. The smart contract terminates when the variable is out of its scope. Then, different instantiation methods are proposed. Finally, we perform the qualitative comparison between the proposed approach and the predeployment method, and it demonstrates that the proposed methods optimize the life cycle of the smart contract and save calculation resources.

Blockchain technology has emerged as a novel distributed ledger technology, facilitating data sharing and system management securely and efficiently without interventions from a central authority. However, blockchain technology alone is not suitable for enterprise-class applications, mainly due to the limitations in capacity expansion and verification speed of blockchain systems. This paper proposes a secure and effective construction scheme for blockchain networks to improve performance and address the effective management concerns of blockchain data based on transaction categories. We designed a network link protocol to construct a directed acyclic graph (DAG) blockchain network and used a sharding protocol to divide the DAG blockchain into multiple category shards to process transactions in parallel. We then extensively evaluated our proposed design on local clusters. The experimental results show that our link and shard protocols achieved high throughput and the category-based sharded DAG blockchain demonstrated high scalability.