Overview of the Second Workshop on Scholarly Document Processing

With the ever-increasing pace of research and high volume of scholarly communication, scholars face a daunting task. Not only must they keep up with the growing literature in their own and related fields, scholars increasingly also need to rebut pseudo-science and disinformation. These needs have motivated an increasing focus on computational methods for enhancing search, summarization, and analysis of scholarly documents. However, the various strands of research on scholarly document processing remain fragmented. To reach out to the broader NLP and AI/ML community, pool distributed efforts in this area, and enable shared access to published research, we held the 2nd Workshop on Scholarly Document Processing (SDP) at NAACL 2021 as a virtual event (https://sdproc.org/2021/). The SDP workshop consisted of a research track, three invited talks and three Shared Tasks (LongSumm 2021, SCIVER and 3C). The program was geared towards NLP, information retrieval, and data mining for scholarly documents, with an emphasis on identifying and providing solutions to open challenges. 1 Workshop description Over the past several years and at various venues, the Joint Workshop on Bibliometric-enhanced IR and NLP for Digital Libraries (BIRNDL1) (Cabanac et al., 2020; Mayr et al., 2018), the Allen Institute for AI, USA IBM Research AI, Haifa Research Lab, Israel SRI International, USA ÚFAL, Charles University, Czech Republic Google AI, USA Oak Ridge National Laboratory, USA The Open University, UK GESIS – Leibniz Institute for the Social Sciences, Germany Elsevier, USA Microsoft Research, USA CL-SciSumm Shared Task, and the International Workshop on Mining Scientific Publications (WOSP2) (Knoth et al., 2020) have established themselves as the principal venues for research in scholarly document processing (SDP). However, as these venues are collocated with conferences that are not focused on NLP, current solutions in this domain lag behind modern techniques generated by the greater NLP community. In 2020, the first SciNLP workshop3 was held online at the AKBC 2020 conference; the workshop brought together interested parties in a talk series focused on various aspects of scientific NLP. The first Scholarly Document Processing (SDP) workshop then took place in co-location with the EMNLP 2020 conference as an online workshop (see overview in Chandrasekaran et al. (2020)), and provided a dedicated venue for those working on SDP to submit and discuss their research. Following these successes and the clear appetite for venues to foster discussions around scholarly NLP, SDP 2021 again aimed to connect researchers and practitioners from different communities working with scientific literature and data and created a premier meeting point to facilitate discussions on open problems in SDP. We believe that ACL events are the most appropriate venue for the SDP workshop for three reasons. First, ACL events are the premier venues for the confluence of NLP and ML, and most of the cornerstone tasks in processing scholarly documents are NLP tasks. Improving machine understanding of scholarly semantics embedded in research papers is essential to furthering many tasks and applications in scholarly document processing. Second, the clear practical importance https://philippmayr.github.io/BIRNDL-WS/ https://wosp.core.ac.uk/ https://scinlp.org/