Search results

The importance of the agri-food supply chain in both food production and distribution has made the issue of its development a critical concern. Based on configuration theory and congruence theory, this research investigates the complex impact of supply chain concentration on financial growth in agri-food supply chains.

The cluster analysis and response surface methodology are employed to analyse the data collected from 207 Chinese agri-food companies from 2010 to 2022.

The results indicate that different combination patterns of supply chain concentration can lead to different levels of financial growth. We discover that congruent supplier and customer concentration is beneficial for companies’ financial growth. This impact is more pronounced when the company is in the agricultural production stage of agri-food supply chains. Post-hoc analysis indicates that there exists an inverted U-shaped relationship between the overall levels of supply chain concentration and financial growth.

Our research uncovers the complex interplay between supply chain base and financial outcomes, thereby revealing significant ramifications for agri-food supply chain managers to optimise their strategies for exceptional financial growth.

This study proposes a combined approach of cluster analysis and response surface analysis for analysing configuration issues in supply chain management.

Fine-tuning pre-trained language models (PLMs), e.g. SciBERT, generally require large numbers of annotated data to achieve state-of-the-art performance on a range of NLP tasks in the scientific domain. However, obtaining fine-tuning data for scientific NLP tasks is still challenging and expensive. In this paper, the authors propose the mix prompt tuning (MPT), which is a semi-supervised method aiming to alleviate the dependence on annotated data and improve the performance of multi-granularity academic function recognition tasks.

Specifically, the proposed method provides multi-perspective representations by combining manually designed prompt templates with automatically learned continuous prompt templates to help the given academic function recognition task take full advantage of knowledge in PLMs. Based on these prompt templates and the fine-tuned PLM, a large number of pseudo labels are assigned to the unlabelled examples. Finally, the authors further fine-tune the PLM using the pseudo training set. The authors evaluate the method on three academic function recognition tasks of different granularity including the citation function, the abstract sentence function and the keyword function, with data sets from the computer science domain and the biomedical domain.

Extensive experiments demonstrate the effectiveness of the method and statistically significant improvements against strong baselines. In particular, it achieves an average increase of 5% in Macro-F1 score compared with fine-tuning, and 6% in Macro-F1 score compared with other semi-supervised methods under low-resource settings.

In addition, MPT is a general method that can be easily applied to other low-resource scientific classification tasks.