This paper explores the use of bilingual parallel corpora as a source of lexical knowledge for cross-lingual textual entailment. We claim that, in spite of the inherent difficulties of the task, phrase tables extracted from parallel data allow to capture both lexical relations between single words, and contextual information useful for inference. We experiment with a phrasal matching method in order to: i) build a system portable across languages, and ii) evaluate the contribution of lexical knowledge in isolation, without interaction with other inference mechanisms. Results achieved on an English-Spanish corpus obtained from the RTE3 dataset support our claim, with an overall accuracy above average scores reported by RTE participants on monolingual data. Finally, we show that using parallel corpora to extract paraphrase tables reveals their potential also in the monolingual setting, improving the results achieved with other sources of lexical knowledge. 1

Proceedings of the Workshop

In this paper we investigate the relation between positive and negative pairs in Textual Entailment (TE), in order to highlight the role of contradiction in TE datasets. We base our analysis on the decomposition of Text-Hypothesis pairs into monothematic pairs, i.e. pairs where only one linguistic phenomenon at a time is responsible for entailment judgment and we argue that such a deeper inspection of the linguistic phenomena behind textual entailment is necessary in order to highlight the role of contradiction. We support our analysis with a number of empirical experiments, which use current available TE systems. 1

Recognizing Textual Entailment (RTE) is to detect an important relation between two texts, namely whether one text can be inferred from the other. For natural language processing, especially for natural language understanding, this is a useful and challenging task. We start with an introduction of the notion of textual entailment, and then define the scope of the recognition task. We summarize previous work and point out two important issues involved, meaning representation and relation recognition. For the former, a general representation based on dependency relations between words or tokens is used to approximate the meaning of the text. For the latter, two categories of approaches, intrinsic and extrinsic ones, are proposed. The two parts of the thesis are dedicated to these two classes of approaches. Intrinsically, we develop specialized modules to deal with different types of entailment; and extrinsically, we explore the connection between RTE and other semantic relations between texts.