View on GitHub

SemEval-2021

The 15th International Workshop on Semantic Evaluation

SemEval-2021: Program

SemEval-2021 will be colocated with ACL 2021. All times shown are UTC.

Logistical notes

All SemEval papers can be found in the proceedings.

Thursday, August 5

Access via the *SEM Zoom link. Session Chair: Lun-Wei Ku

Recently, natural language processing (NLP) has had increasing success and produced extensive industrial applications. Despite being sufficient to enable these applications, current NLP systems often ignore the social part of language, e.g., who says it, in what context, for what goals. In this talk, we take a closer look at social factors in language via a new theory taxonomy, and its interplay with computational methods via two lines of work. The first one studies what makes language persuasive by introducing a semi-supervised method to leverage hierarchical structures in text to recognize persuasion strategies in good-faith requests. The second part demonstrates how various structures in conversations can be utilized to generate better summaries for everyday interaction. We conclude by discussing several open-ended questions towards how to build socially aware language technologies, with the hope of getting closer to the goal of human-like language understanding.

Bio: Diyi Yang is an assistant professor in the School of Interactive Computing at Georgia Tech. She is broadly interested in Computational Social Science, and Natural Language Processing. Diyi received her PhD from the Language Technologies Institute at Carnegie Mellon University. Her work has been published at leading NLP/HCI conferences, and also resulted in multiple award nominations from EMNLP, ICWSM, SIGCHI and CSCW. She is named as a Forbes 30 under 30 in Science, a recipient of IEEE AI 10 to Watch, and has received faculty research awards from Amazon, Facebook, JPMorgan Chase, and Salesforce.

Session chair: Guy Emerson

Session chair: Alexis Palmer

Session chair: Alexis Palmer

Best Task/Paper Awards

Friday, August 6

Session chair: Nathan Schneider

This talk brings a psycholinguistic perspective to the questions of what makes a 'good' sentence for a speaker (or NLG system) to produce and what makes a 'good' inference about the world for a listener (or NLU system) to draw from the sentences they encounter. I consider the link between real-world predictability and text likelihood – do the things that speakers choose to say about the world provide a transparent mapping to how the world really is? This talk will introduce experimental evidence that comprehenders expect speakers to mention newsworthy content (namely content that is not highly predictable from world knowledge). For example, comprehenders who are asked to guess what a speaker is going to say next will infer from the mention of the word 'yellow' that the speaker is unlikely to be talking about something prototypically yellow (they anticipate that the speaker is talking about a shirt instead of a banana) and, more generally, they will guess that a sentence contains content that deviates from their real-world priors (they anticipate a description of a newsworthy situation with properties that are rare in the real world). Such findings have implications for the way we use text to infer meaningful facts about the world and the way we evaluate the felicity and sensibility of a text.

Bio: Hannah Rohde is a Reader in Linguistics & English Language at the University of Edinburgh. She works in experimental pragmatics, using psycholinguistic techniques to investigate questions in areas such as pronoun interpretation, referring expression generation, implicature, presupposition, deception, and the establishment of discourse coherence. Her undergraduate degree was in Computer Science and Linguistics from Brown University, followed by a PhD in Linguistics at the University of California San Diego and postdoctoral fellowships at Northwestern and Stanford. She has helped organise the EU-wide "TextLink: Structuring discourse in multilingual Europe" COST Action network and is a recipient of the Philip Leverhulme Prize in Languages and Literatures. Her dream is to one day experience in-person conferences again – to indulge in standing around in overcrowded corridors, talking to interesting people over conference coffee and biscuits!

Session chair: Natalie Schluter

Session chair: Natalie Schluter

Session chair: Nathan Schneider

SemEval-2022 Tasks