03/2021 On March 17th, Project Debater published a paper at Nature which describes the whole system: "An autonomous debating system". Proud to be part of the team!
03/2021 The proud moment of being an advisor: Two papers accepted at NAACL 2021!
02/2021 I was invited to participate in a panel discussion about "Medical Applications of Artificial Intelligence in the Future: What Should We Know?" at MMII Annual Conference 2021.
01/2021 Together with my colleagues, I have a paper "TDMSci: A Specialized Corpus for Scientific Literature Entity Tagging of Tasks Datasets and Metrics" accepted at EACL 2021.
01/2021 I will serve as a standing reviewer of Computational Linguistics (CL) journal.
12/2020 Two workshops I'm co-organizing (ArgMining 2021 and SustaiNLP 2021) accepted at EMNLP 2021.
12/2020 I gave an invited talk "Constructing Leaderboards from Scientific Literature" at Technical University of Darmstadt UKP.
11/2020 I gave a talk "Project Debater: From Lab Research to Product" at Ireland WaiTALK: Path to Entrepreneurship.
10/2020 I have a paper "Fine-grained Information Status Classification Using Discourse Context-Aware BERT" accepted at COLING 2020.
09/2020 I will give an invited talk "Argument Retrieval in Project Debater" at CLEF 2020 Touché workshop.
08/2020 I will serve as an area chair at EACL 2021.
07/2020 I received 2020 WoC "Technical Innovation in Industry" award.
04/2020 I will serve in the standing review committee of TACL.
04/2020 I have a paper "Bridging Anaphora Resolution as Question Answering" accepted at ACL 2020.
03/2020 I release the code for bridging anaphora resolution using bridging embeddings which described in the paper "A deterministic algorithm for bridging anaphora resolution" [link:bridging-resolution]
11/2019 We have two papers related to argumentation mining accepted at AAAI 2020. The first one "Corpus wide argument mining - a working solution" is from the debater project claim/evidence detection team. The second one is titled "End-to-End Argumentation Knowledge Graph Construction". This paper is from an intern project that I was leading from last winter till this spring. The first author Khalid Al Khatib received IBM PhD scholarship and did an internship with us. Code and dataset described in the paper will be available soon.
07/2019 We release the datasets and the code for extracting tuples of task, dataset, metric, score from NLP papers [link:science-result-extractor]
05/2019 Our long paper "Identification of Tasks, Datasets, Evaluation Metrics, and Numeric Scores for Scientific Leaderboards Construction" got accepted at ACL 2019.
11/2018 I'll attend EMNLP 2018 and present the paper "A deterministic algorithm for bridging anaphora resolution".
10/2018 I gave an invited colloquium talk "Computational Argumentation: A personal Journey" at ICL, Heidelberg University.
10/2018 I gave a tutorial about Anaphora Resolution at DRL AI Seminar.
06/2018 I presented the paper "Enhanced Word Representations for Bridging Anaphora Resolution" at NAACL 2018. The new word embedding resource can be downloanded from here.
10/2017 I gave a talk "Information Status: Tasks, Datasets and Models" at Dublin NLP meetup.
09/2017 The datasets used in the paper "Argument Relation Classification Using a Joint Inference Model" can be downloaded from here.
My primary research interest lies in the area of natural language processing (NLP) where I focus on the tasks of language understanding in the discourse level, such as information status recognition, indirect anaphora resolution (bridging resolution). I'm also interested in computational argumentation and information extraction from scientific literature.
Before joining IBM, I was a Postdoc at the department of Computational Linguistics at Heidelberg University, working with Anette Frank. Before that, I did my PhD at HITS (Heidelberg Institute for Theoretical Studies), working with Michael Strube.