Named entity recognition (NER) is the task to identify mentions of rigid designators from text belonging predefined semantic types such as person, location, organization etc. NER always serves foundation for many natural language applications question answering, summarization, and machine translation. Early systems got a huge success in achieving good performance with cost human engineering designing domain-specific features rules. In recent years, deep learning, empowered by continuous...

Data aggregation is an essential operation in wireless sensor network applications. This paper focuses on the data scheduling problem. Based maximal independent sets, a distributed algorithm to generate collision-free schedule for networks proposed. The time latency of generated by proposed minimized using greedy strategy. bound 24D + 6 Delta 16, where D diameter and maximum node degree. previous with least has (Delta- 1)R, R radius. Thus our contributes additive factor instead...

Few-shot learning under the <inline-formula><tex-math notation="LaTeX">$N$</tex-math></inline-formula> -way notation="LaTeX">$K$</tex-math></inline-formula> -shot setting (i.e., annotated samples for each of classes) has been widely studied in relation extraction (e.g., FewRel) and image classification Mini-ImageNet). Named entity recognition (NER) is typically framed as a sequence labeling problem where classes are inherently entangled together because number sentence not known advance,...

Text segmentation is a fundamental task in natural language processing that comes two levels of granularity: (i) segmenting document into sequence topical segments (topic segmentation), and (ii) sentence elementary discourse units (EDU segmentation). Traditional solutions to the tasks heavily rely on carefully designed features. The recently proposed neural models do not need manual feature engineering, but they either suffer from sparse boundary tags or cannot well handle issue variable...

Named entity recognition (NER) is the task to identify mentions of rigid designators from text belonging predefined semantic types such as person, location, organization etc. NER always serves foundation for many natural language applications question answering, summarization, and machine translation. Early systems got a huge success in achieving good performance with cost human engineering designing domain-specific features rules. In recent years, deep learning, empowered by continuous...

The recent success of Large Language Models (LLMs) has gained significant attention in both academia and industry. Substantial efforts have been made to enhance the zero- few-shot generalization capabilities open-source LLMs through finetuning. Currently, prevailing approach is instruction-tuning, which trains complete real-world tasks by generating responses guided natural language instructions. It worth noticing that such an may underperform sequence token classification tasks. Unlike text...

Background Knee cartilage is the most crucial structure in knee, and reduction of thickness a significant factor occurrence development osteoarthritis. Measuring allows for more accurate assessment wear, but this process relatively time-consuming. Our objectives encompass using various DL methods to segment knee from MRIs taken with different equipment parameters, building DL-based model measuring grading cartilage, establishing standardized database thickness. Methods In retrospective...

Recent advances in relation extraction with deep neural architectures have achieved excellent performance. However, current models still suffer from two main drawbacks: 1) they require enormous volumes of training data to avoid model overfitting and 2) there is a sharp decrease performance when the distribution during testing shift one domain other. It thus vital reduce requirement explicitly difference transferring knowledge another. In this work, we concentrate on few-shot under adaptation...

Genetic diagnosis of ADPKD has been challenging due to the variant heterogeneity, presence duplicated segments, and high GC content exon 1 in PKD1. In our reproductive center, 40 patients were still genetically undiagnosed or diagnosed without single-nucleotide resolution after testing with a short-read sequencing panel 312 phenotype. A combination long-range PCR long-read approach for PKD1 was performed on these patients. LRS additionally identified 10 pathogenic likely variants, including...

Named entity recognition (NER) aims to recognize mentions of rigid designators from text belonging predefined semantic types, such as person, location, and organization. In this article, we focus on a fundamental subtask NER, named boundary detection, which at detecting the start end boundaries an mention in text, without predicting its type. The detection is essentially sequence labeling problem. Existing methods either suffer sparse tags (i.e., entities are rare nonentities common) or they...

In this paper, we focus on named entity boundary detection, which aims to detect the start and end boundaries of an mention in text, without predicting its type. A more accurate robust detection approach is desired alleviate error propagation downstream applications, such as linking fine-grained typing systems. Here, first develop a novel labeling with pointer networks, where output dictionary size depends input, variable. Furthermore, propose AT-Bdry, incorporates adversarial transfer...

Text segmentation is a fundamental task in natural language processing. Depending on the levels of granularity, can be defined as segmenting document into topical segments, or sentence elementary discourse units (EDUs). Traditional solutions to two tasks heavily rely carefully designed features. The recently proposed neural models do not need manual feature engineering, but they either suffer from sparse boundary tags cannot efficiently handle issue variable size output vocabulary. In light...

Recent neural architectures in sequence labeling have yielded state-of-the-art performance on single domain data such as newswires. However, they still suffer from (i) requiring massive amounts of training to avoid overfitting; (ii) huge degradation when there is a shift the distribution between and testing. In this paper, we investigate problem adaptation for under homogeneous heterogeneous settings. We propose MetaSeq, novel meta-learning approach labeling. Specifically, MetaSeq...

Streamer discharges are formed in a dielectric-barrier discharge used for nonthermal plasma generation. The results of simulation streamer type flue gas mixture is reported. A Monte Carlo done to obtain the transport and appropriate rate coefficients. coefficients calculated from solve conservation equations electron, positive negative ions, together with Poisson's equation. G-factor (radicals produced per 100 eV electrical energy input discharge) obtained Townsend-type higher as compared...

In this paper, we focus on named entity boundary detection, which is to detect the start and end boundaries of an mention in text, without predicting its type. The detected entities are input linking or fine-grained typing systems for semantic enrichment. We propose BdryBot, a recurrent neural network encoder-decoder framework with pointer from given sentence. encoder considers both character-level representations word-level embeddings represent words. way, BdryBot does not require any...

Named entity recognition (NER) is the task to identify text spans that mention named entities, and classify them into predefined categories such as person, location, organization, etc. In recent years, deep learning, empowered by continuous real-valued vector representations semantic composition through nonlinear processing, has been employed in NER systems, yielding stat-of-the-art performance. our TKDE paper, we provide a comprehensive review on existing learning techniques for NER. We...

In sensor networks, the adversaries can inject false data reports from compromising nodes. Previous approaches for filtering reports, notably statistical en-route filtering, adopt a simple strategy grouping nodes that requires redundant groups and thus decrease effectiveness. Worse still, they either suffer threshold problem, which may lead to complete breakdown of security protection when is exceeded, or are dependent on sink stationarity specific routing protocols, cannot work with mobile...

Masking methods are popularly used to defend against power analysis attacks in embedded systems. Apart from attack, there also exists glitch attack when porting the design gate level. In this paper, we firstly divided existing masking into different types according their functions, value and applications. Secondly, compared masked S-box hardware implementation. Finally, proposed AES encryption with 32-bit 128-bit data path The experimental results show that our takes up less resources has...

CFSFDP (clustering by fast search and find of density peaks) is recently developed densitybased clustering algorithm.Compared to DBSCAN, it needs less parameters computationally cheap for its non-iteration.Alex. at al have demonstrated power many applications.However, performs not well when there are more than one peak cluster, what we name as "no peaks".In this paper, inspired the idea a hierarchical algorithm CHAMELEON, propose an extension CFSFDP, E_CFSFDP, adapt applications.In...

Xiaotao Gu, Liyuan Liu, Hongkun Yu, Jing Li, Chen Chen, Jiawei Han. Proceedings of the 2021 Conference North American Chapter Association for Computational Linguistics: Human Language Technologies. 2021.

As the next generation of TV, Ultra High Definition Television (UHDTV) is attracting more and people's attention as it provides a new viewing experience. Considering content delivery, due to lack HD resources, direct question for industry that whether state-of-the-art upscaling algorithms can be utilized upscale current or Full resources UHD, gaining benefit from higher resolution but without losing high quality To investigate this, in this study, we upscaled 720p 1080p sequences UHD by...

For an unfamiliar Application Programming Interface (API), software developers often access the official documentation to learn its usage, and post questions related this API on social question answering (Q&A) sites seek solutions. The captures information about functionality parameters, but lacks detailed descriptions in different usage scenarios. On contrary, discussions APIs Q&A provide enriching usages. Moreover, existing code search engines retrieval systems cannot effectively return...