EACL Workshop on Computational Semantics Beyond Events and Roles

Event Notification Type: 
Call for Papers
Abbreviated Title: 
Monday, 3 April 2017
Eduardo Blanco
Roser Morante
Roser SaurĂ­
Submission Deadline: 
Wednesday, 25 January 2017

(Apologies for multiples postings)

First Call for Papers

Workshop Computational Semantics Beyond Events and Roles (SemBEaR)

Colocated with EACL 2017
Valencia, April 3-4, 2016

***With a Special Session on Negation***


During the last decade, semantic representation of text has focused on extracting propositional meaning, i.e., capturing who does what to whom, how, when and where. Several corpora are available, and existing tools extract this kind of knowledge, e.g., role labelers trained on PropBank or NomBank. Nevertheless, most current representations tend to disregard significant meaning encoded in human language. For example, sentences (1) and (2) below share the same argument structure regarding verb contracted, but do not convey the same overall meaning. While in the first example John contracting the disease is factual, in the second it is not:

(1) John likely contracted the disease when a mouse bit him in the Adirondacks.
(2) John never contracted the disease although a mouse bit him in the Adirondacks.

In order to truly capture what these sentences mean, aspects of meaning that go beyond identifying events and their roles (e.g., uncertainty, negation and attribution) must be taken into account.

The Workshop on Computational Semantics Beyond Events and Roles focuses on a broad range of semantic phenomena that lays beyond the identification and linking of eventualities and their semantic arguments with relations such as agent (who), theme (what) and location (where), here so called SemBEaR.

SemBEaR is pervasive in human language and, while studied from a theoretical perspective, computational models are still scarce. Humans use language to describe events that do not correlate with a real situation in the world. They express desires, intentions and plans, and also discuss events that did not happen or are unlikely to happen. Events are often described hypothetically, and speculation can be used to explain why something is a certain way without a strong commitment.
Humans do not always (want to) tell the (whole) truth: they may use deception to hide lies. Devices such as irony and sarcasm are employed to play with words so that what is said is not what is meant. Finally, humans not only describe their personal views or experiences, but also attribute statements to others. These phenomena are not exclusive of opinionated texts. They are ubiquitous in language, including scientific works (Hyland 1998) and news as exemplified below:

(3) Female leaders might have avoided world wars.
(4) Political experts speculate that Donald Trump's meltdown is beginning.
(5) Infected people typically don't become contagious until they develop symptoms.
(6) You can only catch Ebola from coming into direct contact with the bodily fluids of someone who has the disease and is showing symptoms.
(7) We have never seen a human virus change the way it is transmitted.
(8) There is no reason to believe that Ebola virus is any different from any of the viruses that infect humans and have not changed the way that they are spread.
(9) The government did not release the files until 1998.

The SemBEaR Workshop is the continuation of the workshop on Negation and Speculation in Natural Language Processing, collocated with ACL-2010, and the workshops on Extra-Propositional Aspects of Meaning in Computational Linguistics, collocated with ACL-2012, NAACL-2015 and COLING-2016.

In its 2017 edition, the SemBEaR workshop aims at bringing together scientists working on these type of semantic phenomena within computational semantics.
The goal is to attract researchers interested in theoretical frameworks, annotation schemas, modeling and implementing real systems, as well as analyzing the impact of SemBEaR in NLP applications. The workshop also aims at building a bridge between theoretical and computational linguistics. In particular, it will address these topics:

- Negation: verbal/non-verbal, analytic/synthetic, clausal/subclausal; scope and focus
- Modality: defining and annotating types for computational linguistics
- Factuality: determining factuality changes within and across documents
- Veridicity and veridicality: measuring author commitment
- Attribution and perspectives: determining whose perspective is being presented
- Irony and sarcasm
- SemBEaR and lexical resources
- SemBEaR at the sentence and discourse level: how much context is necessary?
- SemBEaR across domains: news, scientific texts, legal documents, economy texts, etc.
- SemBEaR and implicit meaning: what do sentences really mean?
- SemBEaR in spoken language
- (Automatically) extracting SemBEaR: strategies, resources and algorithms
- Supervised, unsupervised and rule-based approaches to extract SemBEaR
- Integrating SemBEaR in the NLP pipeline
- SemBEaR for NLP applications: does it help?

Special Session on Negation

Besides long papers presenting novel research on the topics aforementioned, the workshop will include an annotation exercise focused on negation, as a follow-up of work initiated in the 2016 ExProM workshop. Our goal is twofold:

(1) to work towards a standard for annotating and processing negation,
(2) to foster an international collaboration network.

The Special Session will be a hands-on and discussion session. We solicit two types of papers, that will be the basis for the discussion session:

(i) Position papers from researchers working on negation per se, as well as on NLP applications that need to be sensitive to this level of information, such as machine translation, textual entailment, opinion mining, biomedical information extraction, summarization or language generation. Ideally, these short papers should be written from the perspective of a specific NLP application. As for the content, they should describe the negation phenomena that hinder the application performance, with examples that can be used in a discussion session aimed at finding solutions.

(ii) Position papers where an annotation task related to negation is defined. The paper should describe an (original) annotation task and motivate why it is useful. Ideally, authors should report on a pilot annotation experiment where the task has been tested. The pilot experiment can be done on a few samples of text.

Position papers will present proposals from researchers interested in fundamental research on negation, as well as desiderata to improve performance of NLP applications. During the workshop, participants will discuss existing approaches to annotate and process negation, and what is needed in order to make NLP applications perform well with negated statements.

Participants will discuss annotation schemas and annotate sample sentences provided by the organizers. Sentences from corpora from narratives as well as social media will be provided to participants. While the organizers will lead and coordinate the annotation effort and discussion, we plan to redefine the exercise based on participant feedback and the needs of researchers working on NLP applications.

The papers of the Special Session will not be presented in a traditional session, but their content will be discussed. In order to guarantee a productive working session, workshop participants will be provided one month before the conference with all papers accepted for the Special Session, as well as with the annotation samples reported in these papers and a list of discussion items. The organisers will lead a discussion based on this content.

For the annotation task, participants can use their own datasets or the samples of text provided by the organisers. Samples from different genres and domains will be provided: news reports, opinionated texts, image captions, and biomedical abstracts. Participants who do not use the samples provided by the organisers will be asked to release the annotation sample so that the other participants can prepare for the discussion session.

As a result of the Special Session, the organisers will invite interested participants to participate in a collaboration network enabling further development of the concepts and conclusions resulting from the discussion.

Some annotation tasks that can be addressed are the following:

- Focus of negation
- Implicit positive meanings of negated statements
- Negation in modelling perspectives
- Categories of negated statements that are meaningful for NLP applications
- Affixal negation
- Indirect negation
- Lexical negation
- Quantifiers and negation
- Paraphrasing negated propositions


For the Regular Paper Presentation Session authors are invited to submit papers describing original, unpublished work in the topic areas listed above. Full papers should not exceed eight pages. Additionally, authors are invited to submit short papers not exceeding 4 pages. Short papers usually describe: a small, focused contribution; work in progress; a negative result; an opinion piece; or an interesting application nugget.

For the Special Session on Negation authors are invited to submit papers from 4 to 8 pages. Authors should include in the papers links to any (annotation) material the paper is based on, if is it different from the material provided by the organisers.

All papers can have up to 2 pages of references. All papers will be peer reviewed and all accepted papers will be published in the workshop proceedings. Authors need to follow the style guidelines of the main conference, which are to be found at http://eacl2017.org/index.php/calls/call-for-papers .

Important Dates

January 25, 2017 (GMT-12): Submission deadline for short and full papers
February 13, 2017: Author notification
February 21, 2017: Camera-ready papers due
April 3 (or 4), 2017: SemBEaR workshop


Eduardo Blanco - University of North Texas
Roser Morante - VU Amsterdam
Roser SaurĂ­ - Oxford University Press


Contact details are available at the workshop website.