FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204155637> ?p ?o ?g. }

• W3204155637 abstract "Paraphrase generation is a longstanding NLP task that has diverse applications on downstream NLP tasks. However, the effectiveness of existing efforts predominantly relies on large amounts of golden labeled data. Though unsupervised endeavors have been proposed to alleviate this issue, they may fail to generate meaningful paraphrases due to the lack of supervision signals. In this work, we go beyond the existing paradigms and propose a novel approach to generate high-quality paraphrases with data of weak supervision. Specifically, we tackle the weakly-supervised paraphrase generation problem by: (1) obtaining abundant weakly-labeled parallel sentences via retrieval-based pseudo paraphrase expansion; and (2) developing a meta-learning framework to progressively select valuable samples for fine-tuning a pre-trained language model BART on the sentential paraphrasing task. We demonstrate that our approach achieves significant improvements over existing unsupervised approaches, and is even comparable in performance with supervised state-of-the-arts." @default.
• W3204155637 created "2021-10-11" @default.
• W3204155637 creator A5013881064 @default.
• W3204155637 creator A5018483554 @default.
• W3204155637 creator A5039584843 @default.
• W3204155637 creator A5044455276 @default.
• W3204155637 creator A5047766916 @default.
• W3204155637 creator A5083826804 @default.
• W3204155637 creator A5086000078 @default.
• W3204155637 date "2021-01-01" @default.
• W3204155637 modified "2023-10-11" @default.
• W3204155637 title "Learning to Selectively Learn for Weakly-supervised Paraphrase Generation" @default.
• W3204155637 cites W1776056560 @default.
• W3204155637 cites W1798130076 @default.
• W3204155637 cites W1861492603 @default.
• W3204155637 cites W1980776243 @default.
• W3204155637 cites W2101105183 @default.
• W3204155637 cites W2121056381 @default.
• W3204155637 cites W2142697503 @default.
• W3204155637 cites W2143927888 @default.
• W3204155637 cites W2154652894 @default.
• W3204155637 cites W2155482025 @default.
• W3204155637 cites W2161374612 @default.
• W3204155637 cites W2250225488 @default.
• W3204155637 cites W2252001469 @default.
• W3204155637 cites W2528321558 @default.
• W3204155637 cites W2531908596 @default.
• W3204155637 cites W2592335154 @default.
• W3204155637 cites W2610935556 @default.
• W3204155637 cites W2752971446 @default.
• W3204155637 cites W2776652360 @default.
• W3204155637 cites W2798362442 @default.
• W3204155637 cites W2890397703 @default.
• W3204155637 cites W2911550516 @default.
• W3204155637 cites W2943997064 @default.
• W3204155637 cites W2947683321 @default.
• W3204155637 cites W2949832505 @default.
• W3204155637 cites W2962939608 @default.
• W3204155637 cites W2962953307 @default.
• W3204155637 cites W2963001247 @default.
• W3204155637 cites W2963018920 @default.
• W3204155637 cites W2963126845 @default.
• W3204155637 cites W2963172122 @default.
• W3204155637 cites W2963223306 @default.
• W3204155637 cites W2963341956 @default.
• W3204155637 cites W2963371670 @default.
• W3204155637 cites W2963403868 @default.
• W3204155637 cites W2963506530 @default.
• W3204155637 cites W2963508788 @default.
• W3204155637 cites W2963697299 @default.
• W3204155637 cites W2964053384 @default.
• W3204155637 cites W2964274690 @default.
• W3204155637 cites W2964292098 @default.
• W3204155637 cites W2964309657 @default.
• W3204155637 cites W2966746916 @default.
• W3204155637 cites W2970419266 @default.
• W3204155637 cites W2972136110 @default.
• W3204155637 cites W2978625989 @default.
• W3204155637 cites W2982399380 @default.
• W3204155637 cites W2998239226 @default.
• W3204155637 cites W3016221549 @default.
• W3204155637 cites W3027879771 @default.
• W3204155637 cites W3034466976 @default.
• W3204155637 cites W3035748261 @default.
• W3204155637 cites W3092571094 @default.
• W3204155637 cites W3094596325 @default.
• W3204155637 cites W3099309639 @default.
• W3204155637 cites W3104126699 @default.
• W3204155637 cites W3105982656 @default.
• W3204155637 cites W3110687497 @default.
• W3204155637 cites W3155393281 @default.
• W3204155637 cites W3187984161 @default.
• W3204155637 cites W3212575067 @default.
• W3204155637 doi "https://doi.org/10.18653/v1/2021.emnlp-main.480" @default.
• W3204155637 hasPublicationYear "2021" @default.
• W3204155637 type Work @default.
• W3204155637 sameAs 3204155637 @default.
• W3204155637 citedByCount "1" @default.
• W3204155637 countsByYear W32041556372022 @default.
• W3204155637 crossrefType "proceedings-article" @default.
• W3204155637 hasAuthorship W3204155637A5013881064 @default.
• W3204155637 hasAuthorship W3204155637A5018483554 @default.
• W3204155637 hasAuthorship W3204155637A5039584843 @default.
• W3204155637 hasAuthorship W3204155637A5044455276 @default.
• W3204155637 hasAuthorship W3204155637A5047766916 @default.
• W3204155637 hasAuthorship W3204155637A5083826804 @default.
• W3204155637 hasAuthorship W3204155637A5086000078 @default.
• W3204155637 hasBestOaLocation W32041556371 @default.
• W3204155637 hasConcept C111472728 @default.
• W3204155637 hasConcept C119857082 @default.
• W3204155637 hasConcept C138885662 @default.
• W3204155637 hasConcept C154945302 @default.
• W3204155637 hasConcept C162324750 @default.
• W3204155637 hasConcept C187736073 @default.
• W3204155637 hasConcept C204321447 @default.
• W3204155637 hasConcept C2779530757 @default.
• W3204155637 hasConcept C2780451532 @default.
• W3204155637 hasConcept C2780922921 @default.
• W3204155637 hasConcept C41008148 @default.
• W3204155637 hasConceptScore W3204155637C111472728 @default.

• next