]> 2019-02-28T13:10:48Z Vered Shwartz Building meaningful phrase representations is challenging because phrase meanings are not simply the sum of their constituent meanings. Lexical composition can shift the meanings of the constituent words and introduce implicit information. We tested a broad range of textual representations for their capacity to address these issues. We found that as expected, contextualized word representations perform better than static word embeddings, more so on detecting meaning shift than in recovering implicit information, in which their performance is still far from that of humans. Our evaluation suite, including 5 tasks related to lexical composition effects, can serve future research aiming to improve such representations. Ido Dagan 2019-02-27T16:16:37Z [1902.10618] Still a Pain in the Neck: Evaluating Text Representations on Lexical Composition 2019-05-19T13:47:16Z Vered Shwartz Still a Pain in the Neck: Evaluating Text Representations on Lexical Composition 1902.10618 How well do contextualized word embeddings address lexical composition? They are good in recognizing meaning shift ("give in" is different from "give") but much worse with revealing implicit meaning ("hot tea" is about temperature, "hot debate" isn't). 2019-02-28 2019-02-28 2019-02-28T13:17:25Z Article recommendation with Personalized PageRank and Full Text Search 2019-02-28T17:48:31Z W3C Workshop on Web Standardization for Graph Data 2019-02-28