Can We Trust the Judges? This is the question we asked in validating factuality evaluation methods via answer perturbation. Check out the results at the #EvalLLM2025 workshop at #TALN2025. Blog: Watch: Play:

RT @GwenComte: Congratulations to Youssra REBBOUD who successfully defended her PhD thesis today, supervised by @rtroncy @PasqLisena @maraz….

Thanks to the jury members for having evaluated this work: @elen2016 Enrico Motta @serena_villata @paolopapotti / co supervised with @PasqLisena

The final contribution is a rule-based system that again makes use of fine-grained causal relationships between events for explaining fact-checking results. Evaluated on the #FEVEROUS and #AVeriTeC datasets.

A third contribution is a method that relies on the JoinGT system but incorporates fine-grained causal relationships between events from FARO for generating narratives and stories of events. We demonstrated that qualitatively, the generated summaries are more complete.

This dataset can then be used to fine-tune language models (RoBERTa and REBEL) to be more effective at detecting fine-grained causal relationships between events in texts. A demonstrator making use of these models is available at

A second major contribution is the largest dataset containing fine-grained annotations of causal relationships between events (enable, prevent, intend, directly cause). This dataset comes from common sense (ATOMIC), news corpus (CNC) and synthetic #LLM generated data.

Youssra also investigated how capable #LLMs are for reverse-engineering competency questions and deriving user stories out of formal ontologies. Results demonstrate that recent large models such as GPT4 are effective for general domain.

A first contribution is the Facts and Events Relationships Ontology (FARO) that harmonizes different models representing event relations. This is available at

This is #PhD defense day at @EURECOM . Youssra Rebboud is defending her 3.5 years research work on causal event relationship modeling, extraction and applications derived from them,

RT @pgroth: Trip Report: ESWC 2025 #eswc2025.

RT @eswc_conf: We are pleased to announce that Raphaël Troncy @rtroncy is going to be one of our keynotes! #ESWC2025.Title “Building Knowle….

Odeuropa @ World Expo2025 in Osaka, … blogpost summarizing our journey with @odeuropa by @IngerLeemans.#EXPO2025 #scent.

L’IA face au climat : solution ou partie du problème ?.Cet article revient sur notre participation victorieuse au AI Frugal Challenge en marge du AI Summit sponsorisé par @huggingface @QuotaClimat et au nouveau projet #ClimateSense @paolopapotti @EURECOM.

We had a blast yesterday at @expo2025_japan presenting the @odeuropa research results at the EU Pavilion! Come smell with us at 15h for the last session!.@EURECOM

📢A unique fully funded three-years PhD position is open at POLITO and EURECOM. Title: Enhancing Educational Storytelling with Human-Centered AI in the LLM Era.Application deadline: April 28, 2025.Details:

Starter pack : comment créer sa figurine sur chatGPT, avec quel prompt ? - La Voix du Nord

Thanks to the jury members who have reviewed this work @kbontcheva @serena_villata @em_alam and @dugelay2eurecom. Work supervised with @paolopapotti.

Learn more in these various publications, several of them having been awarded as Best Paper. All code available at and

A third contribution is a novel method to detect a fine-grained notion of similarity between documents (based on entities, concepts, narrative and implication) or across document length (generic, specific, hybrid). We compare #SentenceBERT with graph embeddings!