An interview I recently gave to Focus (in Italian) on the future of intelligent personal assistants and the importance of semantics and pragmatics in Natural Language Understanding diag.uniroma1.it/navigli/pdf/in… #NLP #understanding Focus

The videos of the talks of the 2-day #workshop on 10 Years of #BabelNet held in July are online! mousse-project.org/events/event-a… + BabelNet 5.2 out soon: new resources & updates! Co-sponsors: Ricerca Sapienza European Research Council (ERC) Babelscape AI Journal SapienzaNLP Accademia della Crusca #NLProc

Researcher Roberto Navigli revolutionised multilingual #translation using efficient, sophisticated algorithms & creating big knowledge databases from public sources. 📗 bit.ly/RobertoNavigli 🎧 bit.ly/RobertoNavigli #ERC10kGrantees #ITD2022 Sapienza Università di Roma Ricerca Sapienza

Lots of emotions yesterday during the Agenzia ANSA event on augmenting humanity with #AI. Watch my interview here (in Italian)! I talked about my journey and (a very small part of) the future of NLP SapienzaNLP Babelscape facebook.com/plugins/video.…

📢I will give an AI*IA AIxIA spotlight seminar this Friday at 5pm (CET) on Neuro-Symbolic Multilingual Natural Language Understanding aixia.it/incontri/autum… #AI #NLProc #semantics #SapienzaNLP

Hey #NLProc, looking for a tour of Explicit Multilingual Semantics? At AACL 2022 Roberto Navigli Rexhina Blloshmi Edoardo Barba and Simone Conia hosted a tutorial on recent trends, open problems and state-of-the-art solutions in the field. Link: youtube.com/watch?v=KdGeue…

#NLPaperAlert 📢: Semantic Role Labeling meets Definition Modeling in our new paper in Findings of #emnlp2022. By Simone Conia Edoardo Barba Alessandro Scirè Roberto Navigli European Research Council (ERC) #NLProc Read more 📄: researchgate.net/publication/36…

We reframe #MT evaluation as sequence tagging in our #WMT2022 paper: check MaTESe, an interpretable metric that tags error spans in translations! 📜 statmt.org/wmt22/pdf/2022… By Stefano perrella Lorenzo Proietti Alessandro Scirè Niccolò Campolungo Roberto Navigli Babelscape European Research Council (ERC) #EMNLP2022

Excited about joint work between SapienzaNLP and Babelscape on novel MT evaluation as sequence tagging and on euphemism detection (the latter joint with Carnegie Mellon University and MBZUAI (Mohamed bin Zayed University of Artificial Intelligence)! #NLProc #artificialintelligence

The `rebel-large` model is awesome for relation extraction 🔗 Paired with CUDA, it’s blazing fast ⚡️. With LlamaIndex 🦙 🦙, we can now build a knowledge graph over any text data super quickly! 🕸️ Full Colab notebook showing how you can use it: colab.research.google.com/drive/1G6pcR0p…

📢Happy to share that "Neuralign: A Context-Aware, Cross-Lingual and Fully-Neural Sentence Alignment System for Long Texts" has been accepted to #EACL2024 (main) 🫂Huge thanks to my co-authors Stefan Bejgu Simone Tedeschi Simone Conia Roberto Navigli 📃More details coming soon! #NLProc

Exciting strides in text summarization with LLMs 🚀but verifying their factual accuracy is still an open challenge 🤔 We introduce FENICE, a factuality-oriented metric for summarization with a strong focus on interpretability🔍arxiv.org/abs/2403.02270 #NLProc #LLMs #Factuality

👀 Exciting News! 👀 Happy to announce our latest research paper, “Retrieve, Read and LinK: Fast and Accurate Entity Linking and Relation Extraction on an Academic Budget”, will be presented at #ACL2024! 🚀 Try it out! huggingface.co/spaces/relik-i… Thread below 👇

Check out our 6 papers being presented at #ACL2024! *Today:* NounAtlas: Filling the Gap in Nominal SRL - Semantics 📅12th August @ 16:30 🗣️ Roberto Navigli Language Tech raw.githubusercontent.com/SapienzaNLP/no… 🧵Papers presented in other days in the thread.

🎉 Exciting news! Our paper, "ZEBRA: Zero-Shot Example-Based Retrieval Augmentation for Commonsense Question Answering", has been accepted to #EMNLP2024! ✍️ Work conducted with Simone Conia, Riccardo Orlando and Roberto Navigli at SapienzaNLP 👀 Paper, data and models coming soon!

✨Did you know that 𝐌𝐓 𝐦𝐨𝐝𝐞𝐥𝐬 𝐚𝐧𝐝 𝐋𝐋𝐌𝐬 still 𝐬𝐭𝐫𝐮𝐠𝐠𝐥𝐞 𝐭𝐨 𝐡𝐚𝐧𝐝𝐥𝐞 𝐥𝐞𝐱𝐢𝐜𝐚𝐥 𝐚𝐦𝐛𝐢𝐠𝐮𝐢𝐭𝐲? direct.mit.edu/coli/article/5… #MT #LLMs #NLProc

We've landed in Vienna for #ACL2025! 🇦🇹 Proud to present 4 accepted papers on MT evaluation, coreference resolution, word sense disambiguation & LLM evaluation. Thread below 👇

Even state-of-the-art MT systems struggle with ambiguous words — especially rare senses. DiBiMT is a new fully human-curated benchmark across 8 language pairs to study disambiguation bias in MT. 📄 Read the paper in Computational Linguistics: direct.mit.edu/coli/article/5…

👉 slator.ch/EvaluatingSOTA… Researchers 🔎 from Sapienza University, ETH Zürich, and cohere introduce “translation difficulty estimation,” a method designed to create harder benchmarks for #AI #translation. This new approach promises to help the industry see where AI translation

👉 slator.ch/WMT25Prelimina… 🚨 The WMT25 preliminary results are out. Gemini-2.5-Pro and GPT-4.1 🤖 lead #AI #translation. Commercial engines remain stable but mid-tier. Final rankings (with human evaluation) will be published in November. #xl8 #t9n #LLMs @KocmiTom Eleftherios Avramidis