Information-seeking questions in long-form question answering (LFQA) often prove misleading due to ambiguity or false presupposition in the question.While many existing approaches handle misleading questions, they are tailored to limited questions, which are insufficient in a real-world setting with unpredictable input characteristics.In this work, we propose PreWoMe, a unified approach capable of handling any type of information-seeking question.The key idea of PreWoMe involves extracting presuppositions in the question and exploiting them as working memory to generate feedback and action about the question.Our experiment shows that PreWoMe is effective not only in tackling misleading questions but also in handling normal ones, thereby demonstrating the effectiveness of leveraging presuppositions, feedback, and action for real-world QA settings.

We propose FlowKV, a persistent store tailored for large-scale state management of streaming applications. Unlike existing KV stores, FlowKV leverages information from stream processing engines by taking a principled approach toward exploiting information about how and when the applications access data. FlowKV categorizes data access patterns of window operations according to how window boundaries are set and how tuples inside a window are aggregated, and deploys customized in-memory and on-disk data structures optimized for each pattern. In addition, FlowKV takes window metadata as explicit arguments of read and write methods to predict the moment when a window is read, and then loads the tuples of windows in batches from storage ahead of time. Using the NEXMark benchmark as workload, our experiments show that Apache Flink on FlowKV outperforms Flink on RocksDB or Faster with up to 4.12× throughput gain.

This paper proposes an electric vehicle (EV) scheduling algorithm with fuzzy logic control in smart charging network in order to improve the charging performance of the network significantly. The fuzzy logic control helps that the EV scheduling algorithm determines the proper pair of charging station (CS) and EV logically. The fuzzy inference system determines a weight value by reflecting the multiple charging requirements such as distance between EV and CS, charging time, and charging speed. The weight value describes an EV charging priority, and is used in the scheduling algorithm. The proposed scheduling algorithm focuses on avoiding EV congestion at the CS by reducing the waiting time for charging and balancing charging request rate, which shows how the EV is distributed to the CS with balancing the number of available charging pads. In order to compare the performance of the proposed scheduling algorithm with that of conventional algorithms, the random and max weight scheduling algorithm are also considered. The simulation results show that the proposed scheduling algorithm can improve the performance of the smart charging network in view of waiting time for charging and balancing of charging request rate.

As the number of electric vehicles (EVs) is increasing, a balancing control is needed to decrease the charging delay for EV users and increase the overall number of charged EVs for a charging station (CS) service provider. This paper suggests the local pricing-based decentralized balancing control algorithm to balance the power demand and the number of EVs at each CS. The CSs choose the charging costs according to the power demand, and the probability of sending beacon message is controlled according to the queued ratio. When receiving the beacon messages, each EV chooses the CS with the lowest cost. Simulation results showed that the suggested algorithm could balance the parked pad ratio and the charging power ratio of multiple CSs. Furthermore, the number of completely charged EVs was increased and the quality of service was also increased due to the reduction of charging delay. In conclusion, the suggested algorithm can improve the overall performance at the local area from the perspective of both charging service provider and EV users.

Relay selection and random linear network coding are widely used and can improve the performance of wireless networks significantly. In this paper, random linear network coding which pre-forwards coding coefficients in cooperative communication system is proposed. Single relay selection and multiple relay selection are proposed for the relay selection schemes, which avoid collision problem in relay networks. Moreover, we consider channel correlation between coding coefficient and coded packet for practical environment since they are transmitted separately. We investigate the outage probability and diversity order for random linear network coding with two relay selection schemes.