Most papers are machine-translated to English; some are originally in English.
Showing 12 of 12 papers in Natural Language Understanding And Machine Translation
With the development of financial technology, fintech courses in universities face challenges such as an overemphasis on theory, insufficient practical application, and limited teaching tools, making it difficult for students to master end-to-end investment research operational capabilities. This pa…
We point out that neural networks are not black boxes, and their generalization stems from the ability to dynamically map a dataset to the extrema of the model function. We further prove that the number of extrema in a neural network is positively correlated with the number of its parameters. We the…
[Abstract] This study proposes a method for constructing a prosodic lexicon for Lhasa Tibetan based on a continuous speech database, applicable to low-resource and complex languages. The prosodic lexicon constructed from a small set of high-quality data (3.77 hours, 2,526 sentences) significantly im…
Named Entity Recognition (NER, NamedEntityRecognition) is a critical component in natural language processing systems, widely applied in tasks such as question answering, information retrieval, and relation extraction. Although NER systems have undergone decades of research and development, named en…
Advances in technology and the development of the Internet have enabled the storage and distribution of large-scale unstructured data (such as audio, video, and natural language text). However, any such storage and distribution incurs certain costs, naturally prompting considerations for the efficie…
We point out that neural networks are not black boxes, and their generalization stems from the ability to dynamically map a dataset to the local extrema of the model function. We further prove that the number of local extrema in a neural network is positively correlated with the number of its parame…
This study proposes and validates a novel psycho-judicial attack paradigm termed "Cognitive Trial." Unlike traditional prompt injection, this paradigm is completed through a multi-stage game, exploiting core architectural vulnerabilities in Large Language Models (LLMs) that arise from their pursuit …
Natural language is considered closely intertwined with human cognition, with linguistic structures posited to offer profound insights into the cognitive system. However, as a coding system, natural language encodes diverse objects into unified forms; its prominent formal features capture people’s a…
OpenAI's Generative Pre-trained Transformer 4 (GPT-4) is a powerful large language model with a certain degree of intelligence in understanding and generating coherent text. We are exploring whether GPT-4 is capable of acting as a die, i.e. generating random numbers. We show that GPT-4 does not appe…
Based on comment data from Eastmoney.com's Shenzhen Component Index stock forum between January 1, 2018 and December 31, 2019, this study extracts investor sentiment embedded therein using a deep learning BERT model, and employs a TVP-VAR model to investigate the time-varying dynamic relationships a…
This paper presents a comprehensive survey of subword tokenization methods in neural network-based natural language processing. It first elaborates on the out-of-vocabulary (OOV) problem stemming from closed vocabulary limitations in such approaches, and introduces three prevalent methods for addres…
The emergence and widespread adoption of neural networks have significantly advanced research in pattern recognition and data mining. In recent years, graph neural networks have attracted increasing attention. They have found applications in various domains, including text classification, sequence l…
