Workshop
Registration instructions
IberLEF 2021 will be held online on September 21, 2021, collocated with SEPLN 2021 and CEDI 2021. Click here to register and
In "seleccione el evento principal al que desea inscribirse" select the option "SEPLN-WS-IberLEF" (same price as registering for SEPLN only).
In "seleccione el tipo de inscripción", choose "online" option (125 euros) if you are presenting your work at the workshop. If you are only attending, choose "solo oyente" (40 euros).
The registration form is in Spanish; if you do not speak Spanish, check this help prepared by SEPLN to fill in the form.
Workshop Program
September 21, 2021 (All schedule in CEST time)
1 Tasks parallel sessions (1/3)
11am - 12:30 (CEST): Parallel sessions for DETOXIS, EmoEvalES, VaxxStance, ADoBo and MEDDOPROF (see details below)
2 Tasks parallel sessions (2/3)
12:30-14:00 Parallel sessions for EXIST, MeOffendEs, HAHA and eHealth-KD (see details below)
3 IberLEF Plenary session
[Whova link for the plenary session]
16:00-16:15 Overview of IberLEF 2021
Emotions, Stance and Opinions
16:15 - 16:30 Overview of EmoEvalEs
16:30-16:45 Overview of REST-MEX
16:45-17:00 Overview of VaxxStance
Harmful Information
17:00-17:15 Overview of MeOffendEs
17:15-17:30 Overview of EXIST
17:30-17:45 Overview of DETOXIS
17:45-18:00 Overview of FakeDeS
18:00-18:15 Coffee break
Health-related Information Extraction and Knowledge Discovery
18:15-18:30 eHealth-KD
18:30-18:45 MEDDOPROF
Humour and Irony
18:45-19:00 HAHA
19:00-19:15 IDPT
Lexical Acquisition
19:15-19:30 ADoBo
19:30-19:45 Discussion and wrap-up
3 Tasks Parallel sessions (3/3)
20:30-22:00: REST-MEX (see below)
Detailed Program of the Tasks Parallel Sessions
Tasks Parallel Sessions 1/3 (11:00-12:30 CEST)
DETOXIS Parallel Session
[Whova link for DETOXIS session]
11am-11:10 (CEST): Introduction
11:10-11:30: SINAI at IberLEF-2021 DETOXIS task: Exploring Features as Tasks in a Multi-task Learning Approach to Detecting Toxic Comments. Flor Miriam Plaza-del-Arco, M. Dolores Molina-González, L. Alfonso Ureña-López and M. Teresa Martín-Valdivia
11:30-11:50: GuillemGSubies at IberLEF-2021 DETOXIS task: Detecting Toxicity with Spanish BERT. Guillem García Subies
11:50-12:10: AI-UPV at IberLEF-2021 DETOXIS task: Toxicity Detection in Immigration-Related WebNews Comments Using Transformers and Statistical Models . Angel Felipe Magnossão de Paula and Ipek Baris Schlicht
12:10-12:30: GTH-UPM at DETOXIS-IberLEF 2021: Automatic Detection of Toxic Comments in Social Networks. Sergio Esteban Romero, Ricardo Kleinlein, Cristina Luna-Jiménez, Juan Manuel Montero and Fernando Fernández-Martínez
EmoEvalEs Parallel Session
11:00 – 11:15 Welcome and presentation of the EmoEvalEs task
11:15 – 12:15 Participant's approaches presentations
11:15 - 11:25 GSI-UPM team GSI-UPM at IberLEF2021: Emotion Analysis of Spanish Tweets by Fine-tuning the XLM-RoBERTa Language Model Daniel Vera, Oscar Araque, Carlos A. Iglesias
11:25 - 11:30 Questions 11:30 - 11:40 URJC team URJC-Team at EmoEvalEs 2021: BERT for Emotion Classification in Spanish Tweets Jorge Alberto Flores Sánchez, Soto Montalvo Herranz, Raquel Martínez Unanue
11:40 - 11:45 Questions 11:45 - 11:55 UMU team UMUTeam at EmoEvalEs 2021: Emotion Analysis for Spanish based on Explainable Linguistic Features and Transformers José Antonio García-Díaz, Ricardo Colomo-Palacios, Rafael Valencia-García
11:55 - 12:00 Questions
12:00 - 12:10 WSSC team Emotion Detection for Spanish by Combining LASER Embeddings, Topic Information, and Offense Features Fedor Vitiugin, Giorgio Barnabò
12:10 - 12:15 Questions
12:15 – 12:30 Concluding remarks
16:15 - 16:30 Overview of EmoEvalEs (in plenary session)
MEDDOPROF Parallel Session
11:00-11:30 Overview of the Plan TL/BSC MEDDOPROF Shared Task on Detection and Normalization of Occupations in Spanish Medical Documents
Salvador Lima-López, Barcelona Supercomputing Center, Spain
11:30-11:40 NLNDE at MEDDOPROF: Boosting Transformers in a Low-Resource Setting
Lukas Lange, Bosch Center for Artificial Intelligence, Germany
11:40-11:50 Vicomtech at MEDDOPROF: Automatic Information Extraction and Disambiguation in Clinical Text
Elena Zotova, Vicomtech, Spain
11:50-12:00 TALP at MEDDOPROF: Everything Transformers: Recognition, Classification and Normalisation of Professions and Family Relations
Salvador Medina Herrera, Universitat Politécnica de Catalunya, Spain
12:00-12:10 EdIE-KnowLab at MEDDOPROF: BERT and Approximate String Matching for Automatic Recognition and Normalization of Professions in Spanish Medical Documents
Víctor Suárez-Paniagua, University of Edinburgh, UK
12:10-12:20 SINAI at MEDDOPROF: BERT Representations to Identify Professions and Employment Statuses in Health data
José Alberto Mesa Murgado, Universidad de Jaén, Spain
12:20-12:30 Conclusions and wrap-up on MEDDOPROF and ongoing Plan TL/BSC evaluation efforts
Martin Krallinger, Barcelona Supercomputing Center, Spain
VaxxStance Parallel Session
11:00-12:30:
1 MultiAzterTest@VaxxStance-IberLEF 2021: Identifying Stances with Language Models and Linguistic Features. Itziar Gonzalez-Dios and Kepa Bengoetxea.
2 SQYQP@Vaxxstance: Stance Detection for the Antivaxxers Movement. Jesús Calleja and Ariane Méndez
3 WordUp! at VaxxStance 2021: Combining Contextual Information with Textual and Dependency-Based Syntactic Features for Stance Detection. Mirko Lai, Alessandra Cignarella, Livio Finos, Andrea Sciandra.
4 Discussion/Future Work
ADoBo Parallel Session
11.00-11.15: Introduction: lexical borrowing and ADoBo 2021 shared task
11.20-11.35: The futility of STILTs for the classification of lexical borrowings in Spanish, Javier de la Rosa
11.40-11.55: Invited talk on codeswitching, Constantine Lignos
12.00-12.15: Open discussion: the future of ADoBo
Tasks Parallel Sessions 2/3 (12:30-14:00 CEST)
eHealth-KD Parallel Session
12:30 Task overview (eHealth-KD organisers)
12:50 Vicomtech
13:00 PUCRJ-PUCPR-UFMG
13:10 IXA
13:20 Round of systems' videos
13:30-14:00 Discussion
EXIST Parallel Session
12:30 - 14:00 EXIST Parallel Session
12:30 - 12:35: Welcome and Opening Remarks.
12:35 - 12:50: Sexism Prediction in Spanish and English Tweets Using Monolingual and Multilingual BERT and Ensemble Models. Angel Felipe Magnossão de Paula, Roberto Fray da Silva, Ipek Baris Schlicht.
12:50 - 13:05: Sexism Identification in Social Networks using a Multi-Task Learning System. Flor Miriam Plaza-del-Arco, M. Dolores Molina-González, L. Alfonso Ureña López, M. Teresa Martín-Valdivia.
13:05 - 13:20: Automatic Sexism Detection with Multilingual Transformer Models AIT_FHSTP@EXIST2021. Schütz Mina, Boeck Jaqueline, Liakhovets Daria, Slijepcevic Djordje, Kirchknopf Armin, Hecht Manuel, Bogensperger Johannes, Schlarb Sven, Schindler Alexander, and Zeppelzauer Matthias.
13:20 - 13:35: MultiAzterTest@Exist-IberLEF 2021:Linguistically Motivated Sexism Identification. Kepa Bengoetxea, Itziar Gonzalez-Dios.
13:35 - 13:50: EXIST2021: Detecting Sexism with Transformers and Translation-Augmented Data. Guillem García Subies.
13:50 - 14:00: Discussion.
HAHA Parallel Session
- 12:30: Welcome
- 12:35: Keynote by Professor Rada Mihalcea
- 13:00: Presentation by Jocoso: Karish Grover and Tanishq Goel. "HAHA@IberLEF2021: Humor Analysis using Ensembles of Simple Transformers"
- 13:15: Presentation by team iic: Guillem Garcı́a Subies, David Betancur Sánchez and Alejandro Vaca. "BERT and SHAP for Humor Analysis based on Human Annotation"
- 13:30: Presentation by team RoMa: Mariano Rodríguez, Reynier Ortega-Bueno and Paolo Rosso: "RoMa at HAHA-2021: Deep Reinforcement Learning to Improve a Transformed-based Model for Humor Detection"
- 13:45: Presentation by team UMUTeam: José Antonio García-Díaz and Rafael Valencia-García. "UMUTeam at HAHA 2021: Linguistic Features and Transformers for Analysing Spanish Humor. The What, the How, and to Whom"
MeOffendES Parallel Session
12:30 – 12:35 Introduction to the task
12:35 – 12:45 (+5 min questions) UMUTeam
UMUTeam at MeOffendEs 2021: Ensemble Learning for Offensive Language Identification using Linguistic Features, Fine-grained Negation and Transformers José Antonio García-Díaz, Salud María Jiménez-Zafra, Rafael Valencia-García
12:50 – 13:00 (+5 min questions) NLP-CIC
Evaluation of Intermediate Pre-training for the Detection of Offensive Language Segun Taofeek Aroyehun, Alexander Gelbukh
13:05 – 13:15 (+5 min questions) CIMAT-MTY-GTO
Transformers Pipeline for Offensiveness Detection in Mexican Spanish Social Media Víctor Manuel Gómez Espinosa, Víctor Muñiz Sánchez and Adrián Pastor López Monroy
13:20 – 13:30 (+5 min questions) DCCD-INFOTEC
DCCD-INFOTEC at MeOffendEs@IberLEF21 Subtask 3: A transfer learning approach based on EvoMSA's stacked generalization José J. Calderón and Eric S. Tellez and Mario Graff
13:35 – 13:45 (+5 min questions) CIC-IPN
Using Lexical Resources for Detecting Offensiveness in Mexican Spanish Tweets Daniel Abraham Huerta-Velasco and Hiram Calvo
13:50 – 14:00 Concluding remarks
17:00 – 17:15 Overview of MeOffendEs (at IberLEF main session)
Task Parallel Sessions 3/3 (20:30-22:00 CEST)
Tasks Parallel Sessions 3/3 (20:30-22:00 CEST)
REST-MEX Parallel Session
20:30-22:00 CEST (13:30-15:00 CDT)
20:30-21:00: A recommendation system for tourism based on semantic representations and statistical relational learning. Labsemco-UAEM, Jorge Hermosillo
21:00-21:30: An embeddings based recommendation system for mexican tourism. Alumni-MCE 2GEN, Lizeth Garcia
21:30-22:00: Bert-based Approach for Sentiment Analysis of Spanish Reviews from TripAdvisor, Minería UNAM, Juan Vázquez