NLP-PINGAN-TECH @ CL-SciSumm 2020

Abstract

CL-SciSumm Shared Task at EMNLP 2020 Workshop consists of three subtasks about automatic summarization for research papers. This paper introduces the systems of Task 1A and Task 1B submitted by team NLP-PINGAN-TECH. TASK1A is to identify the cited text spans in the reference paper, and Task 1B is to determine the discourse facet of the cited text. Task 1A is regarded as a binary classification task of sentence pairs and the strategies based on language models are proposed. Integration with contextualized embedding with extra information is further explored in this article. For Task 1B, the pre-trained language models are finetuned to accomplish a multi-label classification task. The results show that extra information can improve the identification of cited text spans. The endto-end trained models outperform models trained with two stages, and the averaged prediction of multi-models is more accurate than an individual one.

Overview and Insights from the Shared Tasks at Scholarly Document Processing 2020: CL-SciSumm, LaySumm and LongSumm

“…Task 1a. (Table 1)NLP-PINGAN-TECH (Chai et al, 2020) achieve the best result on Task 1a when evaluated using sentence overlaps and ngram overlaps using ROUGE SU4. All top 5 of their runs outperforms other systems.…”

“…, CiteQA (Umapathy et al, 2020), IIITBH-IITP (Reddy et al, 2020), IITP-AI-NLP-ML (Mishra et al, 2020), MLU (Huang and Krylova, 2020), MLUHW (Boltze et al, 2020), UniHD (Aumiller et al, 2020), NLP-PINGAN-TECH(Chai et al, 2020) Following teams submitted systems for evaluation on Task 2 also which is an optional bonus task: AUTH, CIST(Li et al, 2020), IIITBH-IITP (Reddy et al, 2020), IITP-AI-NLP-ML(Mishra et al, 2020) …”