News

5th CODI workshop

5th CODI workshop

We’re happy to announce that the 5th edition of the CoDi workshop will take place during EACL (March 17-22) 2024.

The last ten years have seen a dramatic improvement in the ability of NLP systems to understand and produce words and sentences. This development has created a renewed interest in discourse problems as researchers move towards the processing of long-form text and conversations. There is a surge of activity in discourse parsing, coherence models, text summarization, corpora for discourse level reading comprehension, and discourse related/aided representation learning, to name a few. At this juncture, we have organized two Workshops on Computational Approaches to Discourse (CODI) at EMNLP 2020 and at EMNLP 2021 to bring together discourse experts and upcoming researchers. These workshops have catalyzed the speed and knowledge needed to solve such problems and have served as a forum for the discussion of suitable datasets and reliable evaluation methods.

The 5th CODI workshop is planned as a 1 day event which brings together different subcommunities. It will feature 2 invited talks: Anna Rohde and Manfred Stede.

All the information on the website: https://sites.google.com/view/codi2024/ .

ANR AnDiAMO

I’m pleased to announce that the ANR AnDiAMO has been accepted!

AnDiAMO stand for Analysing Discourse Automatically, with Multiple Objectives. The main goal is to build high-performing, robust discourse parsers, while relying on a different, but related objective such as a different language, or another task (eg. POS tagging, temporal analysis) or a downstream application (eg. argument mining, sentiment analysis).

The projects starts this year, and should last 3 years. I’ll be soon seeking for a research engineer and a postdoc!

3rd CODI Workshop

3rd CODI Workshop

We’re happy to announce that the 3rd edition of the CoDi workshop will take place during COLING (Oct 12-17) 2022.

The last ten years have seen a dramatic improvement in the ability of NLP systems to understand and produce words and sentences. This development has created a renewed interest in discourse problems as researchers move towards the processing of long-form text and conversations. There is a surge of activity in discourse parsing, coherence models, text summarization, corpora for discourse level reading comprehension, and discourse related/aided representation learning, to name a few. At this juncture, we have organized two Workshops on Computational Approaches to Discourse (CODI) at EMNLP 2020 and at EMNLP 2021 to bring together discourse experts and upcoming researchers. These workshops have catalyzed the speed and knowledge needed to solve such problems and have served as a forum for the discussion of suitable datasets and reliable evaluation methods.

The 3rd CODI workshop is planned as a 1 1/2 to 2 day event which brings together different subcommunities. It will feature invited talks and regular papers on the first day. The second day will be dedicated to one or more shared tasks and special sessions which focus on the issues mentioned above. After a first successful iteration in 2021 the shared task on coreference resolution (encompassing entity and abstract coreference and bridging anaphora) in dialogue will be held again in 2022 (organized by Lori Levin, Ramesh Manuvinakurike, Vincent Ng, Massimo Poesio, Carolyn Rosé, Michael Strube).

All the information on the website: https://sites.google.com/view/codi-2022.

2nd CoDi Workshop

2nd CoDi Workshop

We’re happy to announce that there will be a second edition of the workshop in 2021, and that it will be a 2-days workshop!

The Workshop on Computational Approaches to Discourse brings together researchers interested in all aspects of discourse and its computational modeling. We welcome symbolic and probabilistic approaches, corpus development and analysis, as well as machine and deep learning approaches to discourse. We appreciate theoretical contributions as well as practical applications, including demos of systems and tools. The goal of the workshop is to provide a forum for the community of NLP researchers working on all aspects of discourse.

This year, there will be two shared tasks associated with the workshop:

The workshop will take place during EMNLP 2021.

More information on: https://sites.google.com/view/codi-2021/accueil

1st CoDi workshop

1st CoDi workshop

We’re delighted to announce that the workshop proposal has been accepted!

The Workshop on Computational Approaches to Discourse brings together researchers interested in all aspects of discourse and its computational modeling. We welcome symbolic and probabilistic approaches, corpus development and analysis, as well as machine and deep learning approaches to discourse. We appreciate theoretical contributions as well as practical applications, including demos of systems and tools. The goal of the workshop is to provide a forum for the community of NLP researchers working on all aspects of discourse.

The workshop will take place during EMNLP 2020, on Nov 11 or 12.

More information on: https://codi-workshop.github.io/

ANR Quantum

Our ANR grant QUANTUM has been accepted!!

QUANTUM stands for: QUestion generAtioN for Textual Understanding via Machine reading. The aim of the project is to improve question-answerinf for unstructured input text. We will focus on how to generate questions whose answer is distributed in the text andwhere the words that make up the question are not necessarily present in that text.

The project’s consortium consists of two academic research centers (IRIT and LORIA) and one industrial partner (Synapse). This project is lead by Tim Van de Cruys, researcher at IRIT (Toulouse).

ToNy won the DisRPT shared task

ToNy won the DisRPT shared task

Our system ToNy (Toulouse Nancy) won the shared task on discourse segmentation within a multi-lingual and multi-frameworks setting!

This is a joint work with Philippe Muller and Mathieu Morey.

We thank the organizers of the DisRPT workshop for proposing this nice shared task, and helping in producing reproducible results.

You can find the paper here and the code here.

Thème : Overlay par Kaira. Texte supplémentaire
Le Cap, Afrique du sud