🧑‍🔬 Francesco Orabona

🏛️ Affiliations

🎓King Abdullah University of Science and TechnologyJeddah · Saudi Arabia2023–2026 · 19 paperscurrent🔬Idiap Research InstituteMartigny-Combe · Switzerland2007–2010 · 16 papers🎓Boston UniversityBoston · United States2018–2023 · 14 papersEmployment · Associate Professor, Department of Electrical & Computer Engineering🎓Toyota Technological Institute at ChicagoChicago · United States2012–2014 · 14 papers🎓Stony Brook UniversityStony Brook · United States2016–2019 · 13 papers🎓University of GenoaGenoa · Italy2005–2007 · 11 papers🎓University of MilanMilan · Italy2010–2011 · 8 papers🏢YahooSunnyvale · United States2014–2019 · 6 papers🏢YahooLondon · United Kingdom2016 · 3 papers🎓École Polytechnique Fédérale de LausanneLausanne · Switzerland2008–2010 · 2 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 3 more institutions on single bylines are not listed.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. Mix, MinHash, and Match: Cross-Source Agreement for Multilingual Pretraining Datasets (2025 · 1 citation)
   Web corpora for languages other than English are crawled again and again by separate groups, so the same pages often appear in several public datasets.
2. Cards Against Contamination: TCG-Bench for Difficulty-Scalable Multilingual LLM Reasoning (2026)
   Game benchmarks for LLMs usually rely on well-known games such as chess, whose strategies are widely documented online.
3. SmolKalam: Ensemble Quality-Filtered Translation at Scale for High Quality Arabic Post-Training Data (2025)
   Arabic LLMs have reasonable pre-training corpora, but large post-training sets are scarce. Such sets need multi-turn chats, reasoning traces and tool calls, and unfiltered machine translation is too noisy for this stage. The authors build SmolKalam, an Arabic version of a curated subset of SmolTalk2. Each sample is…

All works

4 works of theirs in our library, the most cited first; 3 with a page here, the others link their original until we summarise them.

1. From N to N+1: Multiclass Transfer Incremental Learning ↗IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · no summary yetIlja Kuzborskij, Francesco Orabona, Barbara Caputo2013 · 145 citations
2. Mix, MinHash, and Match: Cross-Source Agreement for Multilingual Pretraining DatasetsarXiv:2512.18834Sultan Alrashed, Francesco Orabona2025 · 1 citation
3. Cards Against Contamination: TCG-Bench for Difficulty-Scalable Multilingual LLM ReasoningFindings of the Association for Computational LinguisticsSultan AlRashed, Jianghui Wang, Francesco Orabona2026
4. SmolKalam: Ensemble Quality-Filtered Translation at Scale for High Quality Arabic Post-Training DataarXiv:2511.18411Sultan Alrashed, Chadi Helwe, Francesco Orabona2025

🤝 Co-authors here

2 co-authors with a page, the most papers together first.

• Sultan Alrashed ×3
• Chadi Helwe

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

• Findings of the Association for Computational Linguistics: Findings of the Association for Computational Linguistics 2026
• IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings

🔗 Cited by

Papers of our library cite their works 9 times (of the 149,439 papers whose bibliographies we read); 1 citing paper with a page here →.

Updated 1 November 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/francesco-orabona/