English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Context: Stack Overflow is very helpful for software developers who areseeking answers to programming problems. Previous studies have shown that agrowing number of questions are of low quality and thus obtain less attentionfrom potential answerers. Gao et al. proposed an LSTM-based model (i.e.,BiLSTM-CC) to automatically generate question titles from the code snippets toimprove the question quality. However, only using the code snippets in thequestion body cannot provide sufficient information for title generation, andLSTMs cannot capture the long-range dependencies between tokens. Objective:This paper proposes CCBERT, a deep learning based novel model to enhance theperformance of question title generation by making full use of the bi-modalinformation of the entire question body. Method: CCBERT follows theencoder-decoder paradigm and uses CodeBERT to encode the question body intohidden representations, a stacked Transformer decoder to generate predictedtokens, and an additional copy attention layer to refine the outputdistribution. Both the encoder and decoder perform the multi-headself-attention operation to better capture the long-range dependencies. Thispaper builds a dataset containing around 200,000 high-quality questionsfiltered from the data officially published by Stack Overflow to verify theeffectiveness of the CCBERT model. Results: CCBERT outperforms all the baselinemodels on the dataset. Experiments on both code-only and low-resource datasetsshow the superiority of CCBERT with less performance degradation. The humanevaluation also shows the excellent performance of CCBERT concerning bothreadability and correlation criteria.

Improving Stack Overflow question title generation with copying enhanced CodeBERT model and bi-modal information