Different from precision,fuzziness owns some special features.This paper tries to do a preliminary analysis of previous research on fuzziness and fuzzy expressions in legal texts.
Currently,it is key to convey precise meanings to readers for translators.Observing expressing habits between different languages is the precondition to make target texts more readable.It is crucial to ensure the exac...Currently,it is key to convey precise meanings to readers for translators.Observing expressing habits between different languages is the precondition to make target texts more readable.It is crucial to ensure the exactness and seriousness of legal texts,but it does not mean that translators have to take machined way to leave out and even give up transmitting in-deep cultural implications for the intention of achieving the all-inclusive integrity of target content.It is a correct choice for translators to take overt translation and covert translation in line with traits of source language and target language and differences between civil law system and common law system.The intention of the article is to make an empirical study between overt translation and covert translation.展开更多
深度学习在自然语言处理方面取得了巨大进展,以深度神经网络为代表的模型开始在法律智能判决上被广泛使用。基于Transformer的双向编码器表征法(Bidirectional Encoder Representations from Transformers,BERT)模型能够挖掘法律描述文...深度学习在自然语言处理方面取得了巨大进展,以深度神经网络为代表的模型开始在法律智能判决上被广泛使用。基于Transformer的双向编码器表征法(Bidirectional Encoder Representations from Transformers,BERT)模型能够挖掘法律描述文本中双向上下文信息,利用BERT中自注意力机制完成了罪名预测、法律条款推荐、刑期预测多个司法智能审判任务。为了在长文本案情描述文本上获得更好的效果,进一步解决BERT模型输入文本的长度限制,对于过长的输入文本进行关键信息提取。在文本提取的过程中,充分利用前期训练的基于BERT智能审判模型,对于案情描述中句子的重要性进行评估,提取关键句子减少判断模型的输入长度。将精简后的案情描述文本再送入BERT模型进行司法智能审判学习。相比于直接输入原始案情描述文本的方法,基于文本提取处理后的法律描述在智能审判任务中能够取得更好的效果。展开更多
As any other legal language does,legal English features a wealth of complex legal concepts as well as plenty of highly and unique professional terms and complicated syntax.Proper translation of legal English texts int...As any other legal language does,legal English features a wealth of complex legal concepts as well as plenty of highly and unique professional terms and complicated syntax.Proper translation of legal English texts into legal Chinese ones will throw a great social and economic impact on the society.Accordingly "accuracy" has always been regarded as the primary principle of legal English translation.However,as legal English falls within the ambit common law system and legal Chinese within the civil law system,to obtain real accuracy or even the effect of "equivalence" still remains an ideal pursuit of legal English translation standard.Having probed into Skopostheorie,which "boasts itself of one of the deconstructive translation studies"〔1〕 and focuses on the target-text's function and practicability,the author of this article finds that as to legal English translation skopos theory,in terms of text typology,offers a reasonable elucidation for a couple of translation strategies adopted in the target text.展开更多
文摘Different from precision,fuzziness owns some special features.This paper tries to do a preliminary analysis of previous research on fuzziness and fuzzy expressions in legal texts.
文摘Currently,it is key to convey precise meanings to readers for translators.Observing expressing habits between different languages is the precondition to make target texts more readable.It is crucial to ensure the exactness and seriousness of legal texts,but it does not mean that translators have to take machined way to leave out and even give up transmitting in-deep cultural implications for the intention of achieving the all-inclusive integrity of target content.It is a correct choice for translators to take overt translation and covert translation in line with traits of source language and target language and differences between civil law system and common law system.The intention of the article is to make an empirical study between overt translation and covert translation.
文摘深度学习在自然语言处理方面取得了巨大进展,以深度神经网络为代表的模型开始在法律智能判决上被广泛使用。基于Transformer的双向编码器表征法(Bidirectional Encoder Representations from Transformers,BERT)模型能够挖掘法律描述文本中双向上下文信息,利用BERT中自注意力机制完成了罪名预测、法律条款推荐、刑期预测多个司法智能审判任务。为了在长文本案情描述文本上获得更好的效果,进一步解决BERT模型输入文本的长度限制,对于过长的输入文本进行关键信息提取。在文本提取的过程中,充分利用前期训练的基于BERT智能审判模型,对于案情描述中句子的重要性进行评估,提取关键句子减少判断模型的输入长度。将精简后的案情描述文本再送入BERT模型进行司法智能审判学习。相比于直接输入原始案情描述文本的方法,基于文本提取处理后的法律描述在智能审判任务中能够取得更好的效果。
基金National Social Science Project (2013) of School of Foreign Languages of Southwest University of Political Science and Law : Comparative Studies on English Versions of Laws in the Qing Dynasty,Project Number: 13BYY030
文摘As any other legal language does,legal English features a wealth of complex legal concepts as well as plenty of highly and unique professional terms and complicated syntax.Proper translation of legal English texts into legal Chinese ones will throw a great social and economic impact on the society.Accordingly "accuracy" has always been regarded as the primary principle of legal English translation.However,as legal English falls within the ambit common law system and legal Chinese within the civil law system,to obtain real accuracy or even the effect of "equivalence" still remains an ideal pursuit of legal English translation standard.Having probed into Skopostheorie,which "boasts itself of one of the deconstructive translation studies"〔1〕 and focuses on the target-text's function and practicability,the author of this article finds that as to legal English translation skopos theory,in terms of text typology,offers a reasonable elucidation for a couple of translation strategies adopted in the target text.
