Happy to see paloma paper "Decoding Hate: Exploring Language Models' Reactions to Hate Speech" accepted at main conference at #NAACL2025 🎉📚✨ See you in Albuquerque󠁵󠁳󠁮󠁭󠁿🇺🇸! HYBRIDS

This week Martial Pastor is presenting our paper "Enhancing Discourse Parsing for Local Structures from Social Media with LLM-Generated Data" at #coling2025 📚 aclanthology.org/2025.coling-ma…

🎉🎉 I am thrilled to announce that our short paper "Limitations of Automatic Relevance Assessments with Large Language Models for Fair and Reliable Retrieval Evaluation" (w/ Javier Parapar Alvaro Barreiro) got accepted to #SIGIR2025 See you in Padova! 🇮🇹🍝

Honored to be attending #SIGIR2025 in Padova this summer! Excited to share and get feedback on my research towards Explainable and Safe Systems for Health and connect with the amazing IR community. Can’t wait! 📚 🇮🇹IRLab

Happy to be in beautiful Lucca for #ECIR2025! 🇮🇹 The IRLab is excited to enjoy a fantastic programme 📚 in such an inspiring place 🌞🏛️. Today, Jorge, David Otero, & Eliseo Bao will present their work. On Wednesday, it will be the turn of Anxo and paloma

Listening to Jorge from Universidade da Coruña talk about RAG for Keyphrase Synonym Suggestion. Interesting stuff! Glasgow IR Group UofG Computing Science ECIR2025 #ECIR2025

Today Jorge introduced a RAG-based model for keyphrase synonym suggestion in academic search, and David Otero tackled the power of tests in multiple comparisons in IR at ECIR2025 Code for both papers is available! Time for a coffee break ☕

The IRLab finishes its participation at #ECIR2025 with Eliseo Bao demoing a conversational agent for depression screening, Anxo presenting the eRisk 2025 challenges, and paloma sharing her work on distilling efficient & explainable hate speech detectors

Today we conclude a week of great work in the HYBRIDS project - including training for DCs and insightful exchanges among partners, associated partners, and guests from academia and industry. Thanks to all attendees, and to the staff of CITIC Research UDC, Universidade da Coruña, and HYBRIDS

Just now, paloma presented our paper: Decoding Hate: Exploring Language Models' Reactions to Hate Speech at #NAACL2025! 🧠💬 How do LLMs react to toxic content? Proud of her work! #NLP #LLM #HateSpeechDetection 📄 Paper: aclanthology.org/2025.naacl-lon…

Desde el IRLab queremos agradecer a Enrique Carballo de La Opinión A Coruña por el gran trabajo de difusión y síntesis de nuestros resultados de investigación: laopinioncoruna.es/coruna/2025/05… ¡Gracias por dar visibilidad a la ciencia, esto nos motiva a seguir buscando soluciones!

Hoxe no Día da Ciencia na Rúa o IRLab acompañado dos compañeiros da Universidade da Coruña e CITIC Research UDC para intentar facer da nosa investigación un pasatempo para nenos e maiores e plantar a semente da curiosidade #DCR2025

¡Mil gracias a Lara Vivero y @ondacerocoruna por la entrevista en Más de uno A Coruña! Fue un gustazo hablar sobre el proyecto HYBRIDS , gracias por hacer fácil contar los retos de nuestra investigación. ¡Así da gusto comunicar ciencia! 🧠🎙️ 🔗ondacero.es/emisoras/galic…

🧵 1/ Las redes están llenas de odio. ¿Puede la inteligencia artificial ayudarnos a detectarlo… ⚖️ sin discriminar, 🚫 sin reforzar estereotipos, 🔁 y sin aprender a odiar? Esa es la gran pregunta de mi tesis. 👇 Te lo cuento en este #HiloTesis Crue Universidades

Datos interesantes para aquellos que están ahora eligiendo carrera y tienen la portunidad de elegir la Facultade de Informática da UDC y el Grado de Ingeniería Informatica, grado con el Sello de Excelencia: laopinioncoruna.es/coruna/2025/06…

Automatic relevance assessments with LLMs is a hot topic in the #IR community, so it's no surprise to see David Otero answering tons of questions about his work during the poster session at #SIGIR2025! 🔗dl.acm.org/doi/10.1145/37…

Happy to see paloma work "Personalisation or Prejudice? Addressing Geographic Bias in Hate Speech Detection using Debias Tuning in Large Language Models" accepted at #ICWSM2026 🎉 See you in Los Angeles🇺🇸 w/ Patricia M.Rodilla HYBRIDS 🔗Preprint: arxiv.org/abs/2505.02252