We present models for encoding sentences into embedding vectors that specifically target transfer learning to other NLP tasks. The models are efficient and result in accurate performance on diverse transfer tasks. Two variants of the encoding models allow for trade-offs between accuracy and compute resources. For both variants, we investigate and report the relationship between model complexity, resource consumption, the availability of transfer task training data, and task performance. Comparisons are made with baselines that use word level transfer learning via pretrained word embeddings as well as baselines do not use any transfer learning. We find that transfer learning using sentence embeddings tends to outperform word level transfer. With transfer learning via sentence embeddings, we observe surprisingly good performance with minimal amounts of supervised training data for a transfer task. We obtain encouraging results on Word Embedding Association Tests (WEAT) targeted at detecting model bias. Our pre-trained sentence encoding models are made freely available for download and on TF Hub.

No abstract

No abstract

No abstract

No abstract

No abstract

Purpose This study aims to examine factors affecting restaurant customers’ intention to use near field communication (NFC)-based mobile payment (MP) technology. More specifically, based on the valence theory, this paper examined the impacts of users’ negative valence (perceived risk and privacy concern) and positive valence (utilitarian value and convenience) perceptions toward their NFC-MP technology acceptance. Furthermore, the impacts of individual difference variables (smartphone affinity and compatibility) on users’ negative and positive valences and on their behavioral intentions were analyzed. Design/methodology/approach A self-administered online questionnaire was used to collect the data of the study from 412 restaurant customers. A confirmatory factor analysis (CFA) was used to validate the measurement model. To test the hypothesized model, structural equation modeling (SEM) was used. Findings The study findings demonstrated that privacy concern, utilitarian value and convenience significantly affected individuals’ NFC-MP technology acceptance. In addition, compatibility significantly influenced negative and positive valance constructs and smartphone affinity had a positive impact on positive valance constructs only. Practical implications This study provides valuable practical implications for restaurant operators and hospitality technology vendors in the context of mobile payment systems. Originality/value This study successfully extended the valence framework by adding individual difference constructs to it.

No abstract