Translation is the process of rendering a unit from one language (Source Language) into another (Target Language). When it comes to idioms (fixed expressions consisting two words or more giving meaning different individual words), translators are going face number troubles. This study focuses on translating Arabic that contain HEAD, HEART HHAND. The methodology this based statements collected v...

We introduce a new task of entailment relation aware paraphrase generation which aims at generating conforming to given (e.g. equivalent, forward entailing, or reverse entailing) with respect input. propose reinforcement learning-based weakly-supervised paraphrasing system, ERAP, that can be trained using existing and natural language inference (NLI) corpora without an explicit task-specific co...

In most applications of paraphrasing, contextual information should be considered since a word may have different paraphrases in different contexts. This paper presents a method that automatically acquires lexical contextspecific paraphrases from the web. The method includes two main stages, candidate paraphrase extraction and paraphrase validation. Evaluations were conducted on a news title co...

Two kinds of paraphrases extracted from a bilingual parallel corpus were analyzed. One is from an adjectival predicate sentence to a non-adjectival one. The other is from a passive form to a non-passive form. The ability to extract paraphrases is strongly desired for paraphrasing studies. Although extracting paraphrases from multi-lingual parallel corpora is possible, the type of paraphrases ex...

While substantial studies have been achieved on sentiment analysis to date, it is still challenging to explore enough contextual information or specific cues for polarity classification of short text like online product reviews. In this work we explore review clustering and opinion paraphrasing to build multiple cluster-based classifiers for polarity classification of Chinese product reviews un...

The system we presented for the subtask1 and subtask2 in QAC2 is based on our previous one [12], which utilized a greedy answer seeking model using paraphrasing. We incorporate a re-ranking model for matching questions and passages into the previous system. In the model, we integrated a proximitybased scoring function with the structural-based scoring function. Unfortunately, the result of eval...