面向自然语言处理的对抗攻防与鲁棒性分析综述

随着人工智能技术的飞速发展,深度神经网络在计算机视觉、信号分析和自然语言处理等领域中都得到了广泛应用.自然语言处理通过语法分析、语义分析、篇章理解等功能帮助机器处理、理解及运用人类语言.但是,已有研究表明深度神经网络容易受到对抗文本的攻击,通过产生不可察觉的扰动添加到正常文本中,就能使自然语言处理模型预测错误.为了提高模型的鲁棒安全性,近年来也出现了防御相关的研究工作.针对已有的研究,全面地介绍自然语言处理攻防领域的相关工作,具体而言,首先介绍了自然语言处理的主要任务与相关方法;其次,根据攻击和防御机制对自然语言处理的攻击方法和防御方法进行分类介绍;然后,进一步分析自然语言处理模型的可验证鲁棒性和评估基准数据集,并提供自然语言处理应用平台和工具包的详细介绍;最后总结面向自然语言处理的攻防安全领域在未来的研究发展方向.     

基于支持向量规则的运动控制器自然语言构造方法

本文介绍了一种基于支持向量规则的运动控制器自然语言构造方法，提出利用 支持向量机理论，对通过自然语言构造的模糊控制规则进行支持向量规则抽取，从而获得一 个在指定控制精度下的支持向量规则运动控制器．这种方法可以在给定任务精度下抽取真正 有效的控制规则完成控制任务，使控制规则数及控制器形式得到简化，为未来将基于语言构 造的控制器推向实用奠定了基础．所提控制方法在一个轮式移动机器人系统上进行了语言训 练实验．     

自然语言训练的机器人基本行为控制器

本文探索了一种直接利用自然语言进行机器人运动技能训练的控制方法， 提出了利用模糊神经网络结构作为基本行为控制单元，通过教师的自然语言指令完成针对某 一特定行为的运动经验获取和控制器训练，这是一种更加自然的控制器构造方式，以基本运 动单元为基础，可以进一步实现机器人复杂任务的语言编程与控制．所提控制方法最终在一 个轮式移动机器人系统上进行了语言训练实验．     

自然语言处理评测中的问题与对策

自然语言处理中的评测任务引导和推动着技术、模型和方法上的研究.近年来,新的评测数据集和评测任务不断被提出,与此同时,现有评测暴露的一系列问题也限制了自然语言处理技术的进步.该文从自然语言处理评测的概念、构成、发展和意义出发,分类综述了主流自然语言处理评测的任务和特点,进而总结归纳了自然语言处理评测中的问题及其成因.最后...     

数据库自然语言查询界面

数据库技术的普及使得用户对数据库应用界面的要求越来越高，以往的几类接口都需要用户有较高计算机知识水平，而且必须经过一定的培训，这样就会造成人力物力的浪费而且不利于计算机的普及。本文探讨的是一种更为方便简洁不秀学习即可操作的自然语言界面。     

面向任务的基于深度学习的多轮对话系统与技术

自然语言是人类智慧的结晶,以自然语言的形式与计算机进行交互是人们长久以来的期待。随着自然语言处理技术的发展与深度学习方法的兴起,人机对话系统成为了新的研究热点。人机对话系统按照功能可以分为任务导向型对话系统、闲聊型对话系统、问答型对话系统。任务导向型对话系统是一种典型的人机对话系统,旨在帮助用户完成某些特定的任务,有着十分重要的学术意义和应用价值。文中系统地阐述了一种在实际工程应用中的任务导向型对话系统的通用框架,主要包括自然语言理解、对话管理以及自然语言生成3个部分;介绍了上述各部分所采用的经典深度学习和机器学习方法。最后,对自然语言理解任务进行了实证性的实验验证与分析,结果表明文中内容可以为任务导向型对话系统的构建提供有效指导。     

Using a natural language interface with casual users

Although there is much controversy about the merits of natural language interfaces, little empirical research has been conducted on the use of natural language interfaces for database access, especially for casual users. In this work casual users were observed while interacting with a real-life database using a natural language interface, Intellect.Results show that natural language is an efficient and powerful means for expressing requests. This is especially true for users with a good knowledge of the database contents regardless of training or previous experience with computers. Users generally have a positive attitude towards natural language. The majority of errors users make are directly related to restrictions in the vocabulary. However, feedback helps users understand the language limitations and learn how to avoid or recover from errors. Natural language processing technology is developed enough to handle the limited domain of discourse associated with a database; it is simple enough to support casual users with a general knowledge of the database contents; and it is flexible enough to assist problem-solving behaviour.     

Biocomputing: an insight from linguistics

This paper is placed in a formal framework in which the interdisciplinary study of natural language is conducted by integrating linguistics, computer science and biology. It provides an overview of the field of research, conveying the main biological ideas that have influenced research in linguistics. Our work highlights the main methods of molecular computing that have been applied to the processing and study of the structure of natural language: DNA computing, membrane computing and networks of evolutionary processors. Moreover, some new challenges and lines of research for the future are pointed out, that can provide important improvements in the understanding of natural language as a structure and a human capacity.     

TRUMP: A transportable language understanding program

Transportability has perpetually been the nemesis of natural language processing systems, in both the research and commercial sectors. During the last 20 years, the technology has not moved much closer to providing robust coverage of everyday language, and has failed to produce commercial successes beyond a few specialized interfaces and application programs. the redesign required for each application has limited the impact of natural language systems. Trump (TRansportable Understanding Mechanism Package) is a natural language analyzer that functions in a variety of domains, in both interfaces and text processing. While other similar efforts have treated transportability as a problem in knowledge engineering, Trump instead relies mainly on a “core” of knowledge about language and a set of techniques for applying that knowledge within a domain. the information about words, word meanings, and linguistic relations in this generic knowledge base guides the conceptual framework of language interpretation in each domain. Turmp uses this core knowledge to piece together a conceptual representation of a natural language input by combining generic and specialized inforamtion. the result has been a language processing system that is capable of performing fairly extensive analysis with a minimum of customization for each application.     

An intelligent framework for simulating robot-assisted surgical operations

In this paper, a system based on several intelligent techniques, including speech recognition, natural language processing and linear planning is described. These techniques have been employed to generate a sequence of operations understandable by the control system of a robot that is to perform a semi-automatic surgical task.Thus, a system has been implemented that translates some surgeon's ‘natural’ language into robot-executable commands. A robotic simulator has then been implemented in order to test the planned sequence in a virtual environment.     

面向自然语言处理的预训练技术研究综述

近年来,随着深度学习的快速发展,面向自然语言处理领域的预训练技术获得了长足的进步。早期的自然语言处理领域长期使用Word2Vec等词向量方法对文本进行编码,这些词向量方法也可看作静态的预训练技术。然而,这种上下文无关的文本表示给其后的自然语言处理任务带来的提升非常有限,并且无法解决一词多义问题。ELMo提出了一种上下文相关的文本表示方法,可有效处理多义词问题。其后,GPT和BERT等预训练语言模型相继被提出,其中BERT模型在多个典型下游任务上有了显著的效果提升,极大地推动了自然语言处理领域的技术发展,自此便进入了动态预训练技术的时代。此后,基于BERT的改进模型、XLNet等大量预训练语言模型不断涌现,预训练技术已成为自然语言处理领域不可或缺的主流技术。文中首先概述预训练技术及其发展历史,并详细介绍自然语言处理领域的经典预训练技术,包括早期的静态预训练技术和经典的动态预训练技术;然后简要梳理一系列新式的有启发意义的预训练技术,包括基于BERT的改进模型和XLNet;在此基础上,分析目前预训练技术研究所面临的问题;最后对预训练技术的未来发展趋势进行展望。     

An historical overview of natural language processing systems that learn

A fundamental issue in natural language processing is the prerequisite of an enormous quantity of preprogrammed knowledge concerning both the language and the domain under examination. Manual acquisition of this knowledge is tedious and error prone. Development of an automated acquisition process would prove invaluable.This paper references and overviews a range of the systems that have been developed in the domain of machine learning and natural language processing. Each system is categorised into either a symbolic or connectionist paradigm, and has its own characteristics and limitations described.     

Generating Validation Feedback for Automatic Interpretation of Informal Requirements

This paper describes a method for validating conceptual models of digital systems derived automatically from requirements expressed in natural language. Because natural language is ambiguous and vague, most statements have multiple interpretations. The approach here is to feed back to the requirements authors visualizations of the interpretations of the requirements that have been translated to semantic networks. The visualization task is a component (the Model Generator) of the ASPIN system for automatically interpreting requirements expressed in natural language and diagrams, analyzing the requirements for consistency and completeness, and automatically generating engineering models in the VHDL language. Visualization is performed in two steps: mapping the semantic networks to compound digraphs followed by placement of the nodes of the digraphs to generate a display in terms of icons representing devices, values, actions and events; and connectives indicating carriers, data flow and control dependency.     

Natural Language Processing and User Modeling: Synergies and Limitations

The fields of user modeling and natural language processing have been closely linked since the early days of user modeling. Natural language systems consult user models in order to improve their understanding of users' requirements and to generate appropriate and relevant responses. At the same time, the information natural language systems obtain from their users is expected to increase the accuracy of their user models. In this paper, we review natural language systems for generation, understanding and dialogue, focusing on the requirements and limitations these systems and user models place on each other. We then propose avenues for future research.     

A semantic parsing approach for Bhutanese language of Dzongkha

Developments in the computational analysis of Dzongkha have been limited due to the syntactic complexity of the language. Though the natural language processing domains have witnessed rapid developments over the past decade; very few works has been done in Dzongkha despite of being the national language of Bhutan. In this paper, we have investigated the major problems in Dzongkha processing and have proposed a semantic parsing approach for effective processing of this language. We have used a probabilistic approach and have used the linguistic rules in Dzongkha to remove the ambiguities. Semantic representations along with belief net concepts have been used to increase the accuracy of segmentation, syntactic and semantic analyses. The proposed frame work has been able to solve the major issues related to Dzongkha processing, however needs to be further improved to include all the syntactic variations.     

A linguistic ontology of space for natural language processing

We present a detailed semantics for linguistic spatial expressions supportive of computational processing that draws substantially on the principles and tools of ontological engineering and formal ontology. We cover language concerned with space, actions in space and spatial relationships and develop an ontological organization that relates such expressions to general classes of fixed semantic import. The result is given as an extension of a linguistic ontology, the Generalized Upper Model, an organization which has been used for over a decade in natural language processing applications. We describe the general nature and features of this ontology and show how we have extended it for working particularly with space. Treaitng the semantics of natural language expressions concerning space in this way offers a substantial simplification of the general problem of relating natural spatial language to its contextualized interpretation. Example specifications based on natural language examples are presented, as well as an evaluation of the ontology's coverage, consistency, predictive power, and applicability.     

Computer animated visual translation from natural language to sign language

Computer animation and visualization can facilitate communication between the hearing impaired and those with normal speaking capabilities. This paper presents a model of a system that is capable of translating text from a natural language into animated sign language. Techniques have been developed to analyse language and transform it into sign language in a systematic way. A hand motion coding method as applied to the hand motion representation, and control has also been investigated. Two translation examples are also given to demonstrate the practicality of the system.