For our academic and research staff having a paper accepted to a conference is always something to celebrate, especially so when one of our MSc students has contributed to the research and published a paper as the first author.
As part of our MSc degree programmes students complete a dissertation project over the summer. Dr Chenghua Lin, who is part of our NLP research group, supervised Chengkun Zeng (who studied MSc Data Analytics) for his dissertation project ‘Generating Empathetic Dialogue Responses with Deep Learning Approaches’.
Dr Lin said “I proposed this dissertation project as it aligns with my current research into dialogue systems. Being a research-led university gives students the opportunity to directly contribute to our work. During the dissertation process we decided to write a paper based on the research Chengkun has done for his dissertation and I am delighted that it has been accepted by the INLG conference. Our paper titled Affective Decoding for Empathetic Response Generation received very positive comments from the reviewers, with one reviewer even nominating our paper for the Best Paper Award.”
Chengkun, who is now working for ByteDance, a leading multinational internet technology company, said “The topic of my dissertation is not only cutting-edge and meaningful, but also very challenging. I felt it was not easy to make progress at the beginning. Thanks to Dr. Lin, who gave me solid suggestions to improve the model and experiments, I found a path to make small progress each time, after reading lots of papers and doing many experiments. I am very delighted that we published a paper in a major NLP conference based on my dissertation work. I have learned a lot through the whole process and improved many skills including information searching, programming, mathematical reasoning, experiment design, academic writing, and independent thinking.”
Additional information:
The International Conference on Natural Language Generation (INLG) is organised under the auspices of the Special Interest Group on Natural Language Generation (SIGGEN) of the Association for Computational Linguistics (ACL) and the International Speech Communication Association (ISCA). INLG provides an international forum for presentation and discussion of research on natural language generation, and shapes the future directions of NLG research by soliciting and reviewing high quality, applied and theoretical research findings.