I am a Lecturer (~Assistant Professor) in Natural Language Processing at the Computer Science Department of the University of Sheffield. Before joining the University of Sheffield, I was a postdoctoral researcher working with Prof. Iryna Gurevych at the UKP Lab at the Technical University of Darmstadt. I got my PhD degree from Heidelberg University where I was supervised by Prof. Michael Strube. I have studied at Alzahra University and the Sharif University of Technology in Tehran for my BSc and MSc in computer science.
Reseach Interests
If you are a self-funded Ph.D. student and would like to work with me, I am interested to work on evaluating and improving (1) fairness & robustness, (2) accessibility, (3) reasoning skills, and (4) efficiency of language models. Apart from these, I am interested in coreference resolution and interesting psycholinguistic analyses of text ot multimodal inputs.
News
- Nov: Invited talk at Cambridge NLP seminar series
- Oct: Ajie, my first PhD student, co-supervised with Iryna Gurevych, defended his thesis. I am very proud of him and his work and wish him all the best.
- Oct 2023: joining the EMNLP2023’s Awards Committee for deciding the best papers of the conference
- Oct 2023: Dominic’s paper “Learning From Free-Text Human Feedback – Collect New Datasets Or Extend Existing Ones?”, with Ye Tian, Nikolai Rozanov, Iryna Gurevych is accepted at EMNLP 2023
- Oct 2023: invited talk at OFAI’s (Austrian Research Institute for AI) 2023 Fall Lecture Series
- Oct 2023: Our paper “Scoring Coreference Chains with Split-Antecedent Anaphors”, with my amazing co-authors Massimo, Juntao, and Silviu, is published at the Dialogue and Discourse journal
- Nov 2023: will serve as an area chair for the Lexicon and Semantics track at LREC-COLING 2024
- June 2023: will serve as an area chair of the “Semantics: Lexical, Sentence level, Document Level, Textual Inference, etc.” trackat EMNLP 2023
- June 2023: will serve as a workshop co-chair for EACL 2024
- June 2023: gave a talk on end-to-end reasoning at Bristol
- June 2023: Dominic’s paper on “Arithmetic-Based Pretraining – Improving Numeracy of Pretrained Language Models” is accepted at *SEM
- May 2023: Jasivan’s first PhD work “FERMAT: An Alternative to Accuracy for Numerical Reasoning” is accepted at ACL as a long paper
- April 2023: “The Universal Anaphora Scorer 2.0” with Juntao Yu, Michal Novák, Abdulrahman Aloraini, Silviu Paun, Sameer Pradhan and Massimo Poesiois is accepted at IWCS 2023
- Jan 2023: two papers accepted at EACL: “Transformers with Learnable Activation Functions” with Haishuo Fang, Ji-Ung Lee, and Iryna Gurevych accepted at findings, and “Lessons Learned from a Citizen Science Project for Natural Language Processing” with Jan-Christoph Klie, Ji-Ung Lee, Kevin Stowe, Gözde Şahin, Luke Bates, Dominic Petrak, Richard Eckart de Castilho and Iryna Gurevych accepted at main
- Dec 2022: will serve as an area chair for the ACL2023 “Discourse and Pragmatics” track
- Nov 2022: will be serving as an editorial board member for the Northern European Journal of Language Technology (NEJLT)
- Nov 2022: gave an invited talk about “Learnable Activation Functions in Transformer-based Architectures” at the Computer Science Department of Royal Holloway, University of London (RHUL). The talk was summerizing Transformers with Learnable Activation Functions and Adaptable Adapters papers
- Oct 2022: Layer or Representation Space: What makes BERT-based Metrics Robust? is selected as an outstanding paper at COLING 2022
- Oct 2022: will serve as a senior area chair for the “Ethical and Sustainable NLP” at EACL 2023
- Sep 2022: Our paper Evaluating Coreference Resolvers on Community-based Question Answering: From Rule-based to State of the Art with Haixia Chai, Iryna Gurevych, and Michael Strube is accepted at the CRAC 2022 workshop. The paper ended up at a workshop, but if you are interested in coreference resolution, it is definitely worth a read
- August 2022: Our paper “Layer or Representation Space: What makes BERT-based Metrics Robust?” with Doan Nam Long Vu and Steffen Eger is accepted at COLING 2022
- May 2022: will serve as an area chair for “Efficient Methods for NLP” at EMNLP 2022
- April 2022: Our papers “Adaptable Adapters” with Quentin Delfosse, Kristian Kersting, and Iryna Gurevych and “Falsesum: Generating Document-level NLI Examples for Recognizing Factual Inconsistency in Summarization” with Prasetya Ajie Utama, Joshua Bambrick, and Iryna Gurevych are accepted at NAACL 2022
- April 2022: Our paper “The Universal Anaphora Scorer” with Juntao Yu, Sopan Khosla, Silviu Paun, Sameer Pradhan, and Massimo Poesio is accepted at LREC 2022