The process of reading words depends heavily on efficient visual skills, including analyzing and decomposing basic visual features. Surprisingly, previous reading-related studies have almost exclusively focused on gro...The process of reading words depends heavily on efficient visual skills, including analyzing and decomposing basic visual features. Surprisingly, previous reading-related studies have almost exclusively focused on gross aspects of visual skills, while only very few have investigated the role of finer skills. The present study filled this gap and examined the relations of two finer visual skills measured by grating acuity（the ability to resolve periodic luminance variations across space） and Vernier acuity（the ability to detect/discriminate relative locations of features） to Chinese character-processing as measured by character form-matching and lexical decision tasks in skilled adult readers. The results showed that Vernier acuity was significantly correlated with performance in character form-matching but not visual symbol formmatching, while no correlation was found between grating acuity and character processing. Interestingly, we found no correlation of the two visual skills with lexical decisionperformance. These findings provide for the first time empirical evidence that the finer visual skills, particularly as reflected in Vernier acuity, may directly contribute to an early stage of hierarchical word processing.展开更多
The word processing depth hypothesis implies a positive association between learners＇ word processing and their lexical learning. In research, learners＇ task-inherent involvement load （i.e., word processing） has n...The word processing depth hypothesis implies a positive association between learners＇ word processing and their lexical learning. In research, learners＇ task-inherent involvement load （i.e., word processing） has not been found to be consistently associated with their lexical learning. Meanwhile, existing studies have not obtained consensus results, either, from directly associating learners＇ actual word processing and their lexical learning. Against this backdrop, this paper reports a study investigating the association between Chinese EFL learners＇ actual word processing and their lexical learning in performing a collaborative oral output task. Interactional and statistical analyses revealed that the participants engaged in four types of word processing; their overall word processing was significantly correlated with both their productive and receptive word acquisition and retention; their different types of word processing were significantly correlated with their productive word learning, but showed variances in correlations with their receptive word learning. The findings were discussed from the perspectives of word processing in collaborative output, word processing and lexical learning, and word processing and different modes of lexical learning.展开更多
Background Chinese nonfluent aphasic patients experience apparent speech production deficit, but it remains less known in which part of Chinese speech production this deficit occurs. The present study aimed to examine...Background Chinese nonfluent aphasic patients experience apparent speech production deficit, but it remains less known in which part of Chinese speech production this deficit occurs. The present study aimed to examine the ability of nonfluent aphasic patients in Chinese orthography, phonological and semantic processing via two experiments. Experiment I disclosed the general pattern of deficit of Chinese nonfluent aphasic patients in speech production. Experiment II tested whether this deficit occurs in orthography, phonological or semantic processing. Methods The present study adopted neuropsychological testing methods to compare speech production and Chinese word processing between nonfluent aphasic patients （the patient group） and normal individuals （the control group）. Character reading and word reading tasks were used to test speech production. Chinese radical decision, rhyme decision and semantic decision tasks were used to examine word processing. Reaction time and the correct answer rate were collected. Results The patient group had a longer reaction time and was more prone to errors in both character reading and word reading tasks than was the control group. For the patient group, there was no difference between the reaction time of character reading and word reading, the error rate of the former was higher than the latter. In radical decision task the reaction time and error rate to the radical ＂木＂ were higher in the patient group than in the control group. In the rhyme decision task, the reaction time and error rate to the rhyme ＂ang＂ were higher for the aphasic patients. In the semantic decision task the reaction time to characters in the category of animals was higher for the aphasic patients, yet the error rate was not significantly different between the two groups. Conclusions Nonfluent aphasic patients seemingly have decreased speed of speech production and an increased error rate. There is a deficit in phonological processing of aphasic patients while their semantic processing may remain intact.展开更多
This study investigates how orthographic,semantic and contextual variables—including word length,concreteness,and contextual support—impact on the processing and learning of new words in a second language(L2)when fi...This study investigates how orthographic,semantic and contextual variables—including word length,concreteness,and contextual support—impact on the processing and learning of new words in a second language(L2)when first encountered during reading.Students learning English as a foreign language(EFL)were recruited to read sentences for comprehension,embedded with unfamiliar L2 words that occurred once.Immediately after this,they received a form recognition test,a meaning recall test,and a meaning recognition test.Eye-movement data showed significant effects of word length on both early and late processing of novel words,along with effects of concreteness only on late-processing eye-tracking measures.Informative contexts were read slower than neutral contexts,yet contextual support did not show any direct influence on the processing of novel words.Interestingly,initial learning of abstract words was better than concrete words in terms of form and meaning recognition.Attentional processing of novel L2 words,operationalized by total reading time,positively predicted L2 learners’recognition of new orthographic forms.Taken together,these results suggest:1)orthographic,semantic and contextual factors play distinct roles for initial processing and learning of novel words;2)online processing of novel words contributes to L2 learners’initial knowledge of unfamiliar lexical items acquired from reading.展开更多
One of the critical hurdles, and breakthroughs, in the field of Natural Language Processing (NLP) in the last two decades has been the development of techniques for text representation that solves the so-called curse ...One of the critical hurdles, and breakthroughs, in the field of Natural Language Processing (NLP) in the last two decades has been the development of techniques for text representation that solves the so-called curse of dimensionality, a problem which plagues NLP in general given that the feature set for learning starts as a function of the size of the language in question, upwards of hundreds of thousands of terms typically. As such, much of the research and development in NLP in the last two decades has been in finding and optimizing solutions to this problem, to feature selection in NLP effectively. This paper looks at the development of these various techniques, leveraging a variety of statistical methods which rest on linguistic theories that were advanced in the middle of the last century, namely the distributional hypothesis which suggests that words that are found in similar contexts generally have similar meanings. In this survey paper we look at the development of some of the most popular of these techniques from a mathematical as well as data structure perspective, from Latent Semantic Analysis to Vector Space Models to their more modern variants which are typically referred to as word embeddings. In this review of algoriths such as Word2Vec, GloVe, ELMo and BERT, we explore the idea of semantic spaces more generally beyond applicability to NLP.展开更多
模型劫持攻击是一种新型攻击方式,通过植入特定词语,能够隐蔽地控制模型执行与原始任务截然不同的劫持任务,使模型拥有者的训练算力成本增加的同时面临潜在的法律风险。目前,已有研究针对德-英文语言翻译模型探索了这一攻击方式,但在中...模型劫持攻击是一种新型攻击方式,通过植入特定词语,能够隐蔽地控制模型执行与原始任务截然不同的劫持任务,使模型拥有者的训练算力成本增加的同时面临潜在的法律风险。目前,已有研究针对德-英文语言翻译模型探索了这一攻击方式,但在中文自然语言处理(natural language processing,NLP)领域尚属空白。中文语言的独特性使得其面临不同于其他语言环境的安全挑战,因此亟需开发针对中文模型的攻击评估方法。基于上述事实,提出了一种基于中文逻辑词的模型劫持攻击方法Cheater,用于评估中文模型的安全性。Cheater针对中-英文NLP任务,首先使用公共模型对劫持数据进行伪装生成过渡数据,再通过在过渡样本中嵌入中文逻辑词的方式对其进行改造生成毒性数据,最后利用毒性数据完成对目标模型的劫持。实验表明,对于Bart[large]模型,Cheater在0.5%的数据投毒率下攻击成功率可以达到90.2%。展开更多
基金supported by grants from the National Natural Science Foundation of China (81301175, 31771229 and 31371134)
文摘The process of reading words depends heavily on efficient visual skills, including analyzing and decomposing basic visual features. Surprisingly, previous reading-related studies have almost exclusively focused on gross aspects of visual skills, while only very few have investigated the role of finer skills. The present study filled this gap and examined the relations of two finer visual skills measured by grating acuity（the ability to resolve periodic luminance variations across space） and Vernier acuity（the ability to detect/discriminate relative locations of features） to Chinese character-processing as measured by character form-matching and lexical decision tasks in skilled adult readers. The results showed that Vernier acuity was significantly correlated with performance in character form-matching but not visual symbol formmatching, while no correlation was found between grating acuity and character processing. Interestingly, we found no correlation of the two visual skills with lexical decisionperformance. These findings provide for the first time empirical evidence that the finer visual skills, particularly as reflected in Vernier acuity, may directly contribute to an early stage of hierarchical word processing.
基金supported by the MOE Project of the Center for Linguistics and Applied Linguistics,Guangdong University of Foreign Studies,Chinasupported by China's Educational Ministry humanity social science key research center project(No.12JJD740006)
文摘The word processing depth hypothesis implies a positive association between learners＇ word processing and their lexical learning. In research, learners＇ task-inherent involvement load （i.e., word processing） has not been found to be consistently associated with their lexical learning. Meanwhile, existing studies have not obtained consensus results, either, from directly associating learners＇ actual word processing and their lexical learning. Against this backdrop, this paper reports a study investigating the association between Chinese EFL learners＇ actual word processing and their lexical learning in performing a collaborative oral output task. Interactional and statistical analyses revealed that the participants engaged in four types of word processing; their overall word processing was significantly correlated with both their productive and receptive word acquisition and retention; their different types of word processing were significantly correlated with their productive word learning, but showed variances in correlations with their receptive word learning. The findings were discussed from the perspectives of word processing in collaborative output, word processing and lexical learning, and word processing and different modes of lexical learning.
基金This study was supported by a grant from the National Natural Science Foundation of China （No. 30570647）.
文摘Background Chinese nonfluent aphasic patients experience apparent speech production deficit, but it remains less known in which part of Chinese speech production this deficit occurs. The present study aimed to examine the ability of nonfluent aphasic patients in Chinese orthography, phonological and semantic processing via two experiments. Experiment I disclosed the general pattern of deficit of Chinese nonfluent aphasic patients in speech production. Experiment II tested whether this deficit occurs in orthography, phonological or semantic processing. Methods The present study adopted neuropsychological testing methods to compare speech production and Chinese word processing between nonfluent aphasic patients （the patient group） and normal individuals （the control group）. Character reading and word reading tasks were used to test speech production. Chinese radical decision, rhyme decision and semantic decision tasks were used to examine word processing. Reaction time and the correct answer rate were collected. Results The patient group had a longer reaction time and was more prone to errors in both character reading and word reading tasks than was the control group. For the patient group, there was no difference between the reaction time of character reading and word reading, the error rate of the former was higher than the latter. In radical decision task the reaction time and error rate to the radical ＂木＂ were higher in the patient group than in the control group. In the rhyme decision task, the reaction time and error rate to the rhyme ＂ang＂ were higher for the aphasic patients. In the semantic decision task the reaction time to characters in the category of animals was higher for the aphasic patients, yet the error rate was not significantly different between the two groups. Conclusions Nonfluent aphasic patients seemingly have decreased speed of speech production and an increased error rate. There is a deficit in phonological processing of aphasic patients while their semantic processing may remain intact.
文摘This study investigates how orthographic,semantic and contextual variables—including word length,concreteness,and contextual support—impact on the processing and learning of new words in a second language(L2)when first encountered during reading.Students learning English as a foreign language(EFL)were recruited to read sentences for comprehension,embedded with unfamiliar L2 words that occurred once.Immediately after this,they received a form recognition test,a meaning recall test,and a meaning recognition test.Eye-movement data showed significant effects of word length on both early and late processing of novel words,along with effects of concreteness only on late-processing eye-tracking measures.Informative contexts were read slower than neutral contexts,yet contextual support did not show any direct influence on the processing of novel words.Interestingly,initial learning of abstract words was better than concrete words in terms of form and meaning recognition.Attentional processing of novel L2 words,operationalized by total reading time,positively predicted L2 learners’recognition of new orthographic forms.Taken together,these results suggest:1)orthographic,semantic and contextual factors play distinct roles for initial processing and learning of novel words;2)online processing of novel words contributes to L2 learners’initial knowledge of unfamiliar lexical items acquired from reading.
文摘One of the critical hurdles, and breakthroughs, in the field of Natural Language Processing (NLP) in the last two decades has been the development of techniques for text representation that solves the so-called curse of dimensionality, a problem which plagues NLP in general given that the feature set for learning starts as a function of the size of the language in question, upwards of hundreds of thousands of terms typically. As such, much of the research and development in NLP in the last two decades has been in finding and optimizing solutions to this problem, to feature selection in NLP effectively. This paper looks at the development of these various techniques, leveraging a variety of statistical methods which rest on linguistic theories that were advanced in the middle of the last century, namely the distributional hypothesis which suggests that words that are found in similar contexts generally have similar meanings. In this survey paper we look at the development of some of the most popular of these techniques from a mathematical as well as data structure perspective, from Latent Semantic Analysis to Vector Space Models to their more modern variants which are typically referred to as word embeddings. In this review of algoriths such as Word2Vec, GloVe, ELMo and BERT, we explore the idea of semantic spaces more generally beyond applicability to NLP.
文摘模型劫持攻击是一种新型攻击方式,通过植入特定词语,能够隐蔽地控制模型执行与原始任务截然不同的劫持任务,使模型拥有者的训练算力成本增加的同时面临潜在的法律风险。目前,已有研究针对德-英文语言翻译模型探索了这一攻击方式,但在中文自然语言处理(natural language processing,NLP)领域尚属空白。中文语言的独特性使得其面临不同于其他语言环境的安全挑战,因此亟需开发针对中文模型的攻击评估方法。基于上述事实,提出了一种基于中文逻辑词的模型劫持攻击方法Cheater,用于评估中文模型的安全性。Cheater针对中-英文NLP任务,首先使用公共模型对劫持数据进行伪装生成过渡数据,再通过在过渡样本中嵌入中文逻辑词的方式对其进行改造生成毒性数据,最后利用毒性数据完成对目标模型的劫持。实验表明,对于Bart[large]模型,Cheater在0.5%的数据投毒率下攻击成功率可以达到90.2%。
