FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3111997029> ?p ?o ?g. }

• W3111997029 abstract "The ubiquity of smart voice assistants has made conversational shopping commonplace. This is especially true for low consideration segments like grocery. A central problem in conversational grocery is the automatic generation of short product titles that can be read out fast during a conversation. Several supervised models have been proposed in the literature that leverage manually labeled datasets and additional product features to generate short titles automatically. However, obtaining large amounts of labeled data is expensive and most grocery item pages are not as feature-rich as other categories. To address this problem we propose a pre-training based solution that makes use of unlabeled data to learn contextual product representations which can then be fine-tuned to obtain better title compression even in a low resource setting. We use a self-attentive BiLSTM encoder network with a time distributed softmax layer for the title compression task. We overcome the vocabulary mismatch problem by using a hybrid embedding layer that combines pre-trained word embeddings with trainable character level convolutions. We pre-train this network as a discriminator on a replaced-token detection task over a large number of unlabeled grocery product titles. Finally, we fine tune this network, without any modifications, with a small labeled dataset for the title compression task. Experiments on Walmart's online grocery catalog show our model achieves performance comparable to state-of-the-art models like BERT and XLNet. When fine tuned on all of the available training data our model attains an F1 score of 0.8558 which lags the best performing model, BERT-Base, by 2.78% and XLNet by 0.28% only, while using 55 times lesser parameters than both. Further, when allowed to fine tune on 5% of the training data only, our model outperforms BERT-Base by 24.3% in F1 score." @default.
• W3111997029 created "2020-12-21" @default.
• W3111997029 creator A5013913636 @default.
• W3111997029 creator A5017429163 @default.
• W3111997029 creator A5023707138 @default.
• W3111997029 date "2020-12-13" @default.
• W3111997029 modified "2023-09-25" @default.
• W3111997029 title "Discriminative Pre-training for Low Resource Title Compression in Conversational Grocery." @default.
• W3111997029 cites W1832693441 @default.
• W3111997029 cites W1980570680 @default.
• W3111997029 cites W2151258001 @default.
• W3111997029 cites W2218641061 @default.
• W3111997029 cites W2251654079 @default.
• W3111997029 cites W2251656952 @default.
• W3111997029 cites W2317879529 @default.
• W3111997029 cites W2467173223 @default.
• W3111997029 cites W2551396370 @default.
• W3111997029 cites W2574535369 @default.
• W3111997029 cites W2733964180 @default.
• W3111997029 cites W2782541074 @default.
• W3111997029 cites W2791665887 @default.
• W3111997029 cites W2884900624 @default.
• W3111997029 cites W2888103369 @default.
• W3111997029 cites W2890116189 @default.
• W3111997029 cites W2900174585 @default.
• W3111997029 cites W2902744721 @default.
• W3111997029 cites W2951652470 @default.
• W3111997029 cites W2951777553 @default.
• W3111997029 cites W2963241389 @default.
• W3111997029 cites W2963341956 @default.
• W3111997029 cites W2963620441 @default.
• W3111997029 cites W2963929190 @default.
• W3111997029 cites W2965506557 @default.
• W3111997029 cites W2965854689 @default.
• W3111997029 cites W2996035354 @default.
• W3111997029 cites W3038058348 @default.
• W3111997029 hasPublicationYear "2020" @default.
• W3111997029 type Work @default.
• W3111997029 sameAs 3111997029 @default.
• W3111997029 citedByCount "0" @default.
• W3111997029 crossrefType "posted-content" @default.
• W3111997029 hasAuthorship W3111997029A5013913636 @default.
• W3111997029 hasAuthorship W3111997029A5017429163 @default.
• W3111997029 hasAuthorship W3111997029A5023707138 @default.
• W3111997029 hasConcept C108583219 @default.
• W3111997029 hasConcept C111919701 @default.
• W3111997029 hasConcept C118505674 @default.
• W3111997029 hasConcept C138885662 @default.
• W3111997029 hasConcept C153083717 @default.
• W3111997029 hasConcept C154945302 @default.
• W3111997029 hasConcept C162324750 @default.
• W3111997029 hasConcept C187736073 @default.
• W3111997029 hasConcept C188441871 @default.
• W3111997029 hasConcept C204321447 @default.
• W3111997029 hasConcept C2524010 @default.
• W3111997029 hasConcept C2776401178 @default.
• W3111997029 hasConcept C2777601683 @default.
• W3111997029 hasConcept C2779803651 @default.
• W3111997029 hasConcept C2780451532 @default.
• W3111997029 hasConcept C28490314 @default.
• W3111997029 hasConcept C33923547 @default.
• W3111997029 hasConcept C41008148 @default.
• W3111997029 hasConcept C41608201 @default.
• W3111997029 hasConcept C41895202 @default.
• W3111997029 hasConcept C76155785 @default.
• W3111997029 hasConcept C90673727 @default.
• W3111997029 hasConcept C94915269 @default.
• W3111997029 hasConcept C97931131 @default.
• W3111997029 hasConceptScore W3111997029C108583219 @default.
• W3111997029 hasConceptScore W3111997029C111919701 @default.
• W3111997029 hasConceptScore W3111997029C118505674 @default.
• W3111997029 hasConceptScore W3111997029C138885662 @default.
• W3111997029 hasConceptScore W3111997029C153083717 @default.
• W3111997029 hasConceptScore W3111997029C154945302 @default.
• W3111997029 hasConceptScore W3111997029C162324750 @default.
• W3111997029 hasConceptScore W3111997029C187736073 @default.
• W3111997029 hasConceptScore W3111997029C188441871 @default.
• W3111997029 hasConceptScore W3111997029C204321447 @default.
• W3111997029 hasConceptScore W3111997029C2524010 @default.
• W3111997029 hasConceptScore W3111997029C2776401178 @default.
• W3111997029 hasConceptScore W3111997029C2777601683 @default.
• W3111997029 hasConceptScore W3111997029C2779803651 @default.
• W3111997029 hasConceptScore W3111997029C2780451532 @default.
• W3111997029 hasConceptScore W3111997029C28490314 @default.
• W3111997029 hasConceptScore W3111997029C33923547 @default.
• W3111997029 hasConceptScore W3111997029C41008148 @default.
• W3111997029 hasConceptScore W3111997029C41608201 @default.
• W3111997029 hasConceptScore W3111997029C41895202 @default.
• W3111997029 hasConceptScore W3111997029C76155785 @default.
• W3111997029 hasConceptScore W3111997029C90673727 @default.
• W3111997029 hasConceptScore W3111997029C94915269 @default.
• W3111997029 hasConceptScore W3111997029C97931131 @default.
• W3111997029 hasLocation W31119970291 @default.
• W3111997029 hasOpenAccess W3111997029 @default.
• W3111997029 hasPrimaryLocation W31119970291 @default.
• W3111997029 hasRelatedWork W165710780 @default.
• W3111997029 hasRelatedWork W2134928896 @default.
• W3111997029 hasRelatedWork W2404161646 @default.
• W3111997029 hasRelatedWork W2754977407 @default.
• W3111997029 hasRelatedWork W2911424454 @default.

• next