Question Generation(QG)is the task of generating questions according to the given *** of the existing methods are based on Recurrent Neural Networks(RNNs)for generating questions with passage-level input for providing...
详细信息
Question Generation(QG)is the task of generating questions according to the given *** of the existing methods are based on Recurrent Neural Networks(RNNs)for generating questions with passage-level input for providing more details,which seriously suffer from such problems as gradient vanishing and ineffective information *** fact,reasonably extracting useful information from a given context is more in line with our actual needs during questioning especially in the education *** that end,in this paper,we propose a novel Hierarchical Answer-Aware and Context-Aware Network(HACAN)to construct a high-quality passage representation and judge the balance between the sentences and the whole ***,a Hierarchical Passage Encoder(HPE)is proposed to construct an answer-aware and context-aware passage representation,with a strategy of utilizing multi-hop ***,we draw inspiration from the actual human questioning process and design a Hierarchical Passage-aware Decoder(HPD)which determines when to utilize the passage *** conduct extensive experiments on the SQuAD dataset,where the results verify the effectivenesss of our model in comparison with several baselines.
Machine learning-assisted retrosynthesis planning aims to utilize machine learning(ML)algorithms to find synthetic pathways for target *** recent years,with the development of artificial intelligence(AI),especially ML...
详细信息
Machine learning-assisted retrosynthesis planning aims to utilize machine learning(ML)algorithms to find synthetic pathways for target *** recent years,with the development of artificial intelligence(AI),especially ML,researchers’interest in ML-assisted retrosynthesis planning has rapidly increased,bringing development and opportunities to the *** this review,we aim to provide a comprehensive understanding of ML-assisted retrosynthesis *** first discuss the formal definition and the objective of retrosynthesis planning,and organize a modular framework which includes four modules:data preparation,data preprocessing,pathway generation and evaluation,and pathway ***,we sequentially review the current status of the first three modules(except pathway verification)in the ML-assisted retrosynthesis planning framework,including ideas,methods,and latest *** that,we specifically discuss large language models in retrosynthesis ***,we summarize the extant challenges that are faced by current ML-assisted retrosynthesis planning research and offer a perspective on future research directions and development.
The rapid evolution of unmanned aerial vehicle(UAV)technology and autonomous capabilities has positioned UAV as promising last-mile delivery *** and onboard UAV collaborative delivery is introduced as a novel delivery...
详细信息
The rapid evolution of unmanned aerial vehicle(UAV)technology and autonomous capabilities has positioned UAV as promising last-mile delivery *** and onboard UAV collaborative delivery is introduced as a novel delivery *** collaboration,along with energy consumption with payload and wind conditions play important roles in delivery route *** paper introduces the traveling salesman problem with time window and onboard UAV(TSPTWOUAV)and emphasizes the consideration of real-world scenarios,focusing on time collaboration and energy consumption with wind and *** address this,a mixed integer linear programming(MILP)model is formulated to minimize the energy consumption costs of vehicle and ***,an adaptive large neighborhood search(ALNS)algorithm is applied to identify high-quality solutions *** effectiveness of the proposed model and algorithm is validated through numerical tests on real geographic instances and sensitivity analysis of key parameters is conducted.
Starting from the multi-soliton solutions obtained by the Hirota bilinear method,the soli ton molecule structures for the combined mKdV-type bilinear equation(Dt+∑n=1NαnDx2n+1)f*·f=0 are investigated using the ...
详细信息
Starting from the multi-soliton solutions obtained by the Hirota bilinear method,the soli ton molecule structures for the combined mKdV-type bilinear equation(Dt+∑n=1NαnDx2n+1)f*·f=0 are investigated using the velocity resonance *** two-soliton molecules of the mKdV-35 equation and the three-soliton molecules of the mKdV-357 equation are specifically demonstrated in this *** particular selections of the involved arbitrary parameters,especially the wave numbers,it is confirmed that,besides the usual multi-bright soliton molecules,the multi-dark soliton molecules and the mixed multibright-dark soliton molecules can also be *** addition,we discuss the existence of the multi-soliton molecules for the combined mKdV-type bilinear equation with more higher order nonlinear terms and *** results demonstrate that when N≥4,the combined mKdVtype bilinear equation no longer admits soliton molecules comprising more than four solitons.
This article proposes an innovative maritime 5G communication system based on Software Defined Radio (SDR) and OpenAirInterface (OAI),which shows advantages of high scalability and low cost. By implementing with recen...
详细信息
Maritime communication faces the influence of many irresistible factors such as air pressure and waves, which leads to the degradation of the efficiency and quality of maritime communication, and increases the difficu...
详细信息
In view of the various dangerous situations that ships may face the sea, such as sea wind, waves, and ocean currents, which can cause problems such as tilting, ascending and descending, and swaying of the ship's b...
详细信息
This paper investigates the traffic flow of connected and automated vehicles(CAVs)inducing by a moving bottleneck on a two-lane highway.A heuristic rules-based algorithm(HRA)has been used to control the traffic flow u...
详细信息
This paper investigates the traffic flow of connected and automated vehicles(CAVs)inducing by a moving bottleneck on a two-lane highway.A heuristic rules-based algorithm(HRA)has been used to control the traffic flow upstream of the moving *** the HRA,some CAVs in the control zone are mapped onto the neighboring lane as virtual *** improve the driving comfort,the command acceleration caused by virtual vehicle is *** with the benchmark in which the CAVs change lane as soon as the lane changing condition is met,the HRA significantly improves the traffic flow:the overtaking throughput as well as the outflow rate increases,the travel delay and the fuel consumption decrease,the comfort level could also be improved.
Recently advancements in deep learning models have significantly facilitated the development of sequential recommender systems(SRS).However,the current deep model structures are limited in their ability to learn high-...
详细信息
Recently advancements in deep learning models have significantly facilitated the development of sequential recommender systems(SRS).However,the current deep model structures are limited in their ability to learn high-quality embeddings with insufficient ***,highly skewed long-tail distribution is very common in recommender ***,in this paper,we focus on enhancing the representation of tail items to improve sequential recommendation *** empirical studies on benchmarks,we surprisingly observe that both the ranking performance and training procedure are greatly hindered by the poorly optimized tail item *** address this issue,we propose a sequential recommendation framework named TailRec that enables contextual information of tail item well-leveraged and greatly improves its corresponding *** the characteristics of the sequential recommendation task,the surrounding interaction records of each tail item are regarded as contextual information without leveraging any additional side *** approach allows for the mining of contextual information from cross-sequence behaviors to boost the performance of sequential *** a light contextual filtering component is plug-and-play for a series of SRS *** verify the effectiveness of the proposed TailRec,we conduct extensive experiments over several popular benchmark *** experimental results demonstrate that TailRec can greatly improve the recommendation results and speed up the training *** codes of our methods have been available.
Feature selection methods rooted in rough sets confront two notable limitations:their high computa-tional complexity and sensitivity to noise,rendering them impractical for managing large-scale and noisy *** primary i...
详细信息
Feature selection methods rooted in rough sets confront two notable limitations:their high computa-tional complexity and sensitivity to noise,rendering them impractical for managing large-scale and noisy *** primary issue stems from these methods’undue reliance on all *** overcome these challenges,we introduce the concept of cross-similarity grounded in a robust fuzzy relation and design a rapid and robust feature selection ***,we construct a robust fuzzy relation by introducing a truncation ***,based on this fuzzy relation,we propose the concept of cross-similarity,which emphasizes the sample-to-sample similarity relations that uniquely determine feature importance,rather than considering all such relations *** studying the manifestations and properties of cross-similarity across different fuzzy granularities,we propose a forward greedy feature selection algorithm that leverages cross-similarity as the foundation for information *** algorithm significantly reduces the time complexity from O(m2n2)to O(mn2).Experimental findings reveal that the average runtime of five state-of-the-art comparison algorithms is roughly 3.7 times longer than our algorithm,while our algorithm achieves an average accuracy that surpasses those of the five comparison algorithms by approximately 3.52%.This underscores the effectiveness of our *** paper paves the way for applying feature selection algorithms grounded in fuzzy rough sets to large-scale gene datasets.
暂无评论