Damiano Fornasiere

Artificial Intelligence

Consequence invariance and situational awareness 10th annual Center for Human-Compatible AI (CHAI) workshop. Berkeley, CA, USA
A vocabulary for internal states: how language models name perturbations to their activations MIT & Kempner Institute workshop “Invented, constructed, and emergent languages in multi-agent AI systems”. Cambridge, MA, USA
Can language models recognize dropout and Gaussian noise? Harvard Data Science Initiative & Faculty of Arts and Sciences “Generative AI” workshop. Cambridge, MA, USA (online)
The challenge of evaluation- and training-awareness Mila’s Community of Practice, “AI governance”. Montreal, Canada
Trends and surprises in empirical AI safety Harvard NSAPH seminar. Cambridge, MA, USA (online)

Mathematics

Trees and spectra of Heyting algebras Italian Association of Logic and its Applications (AILA) 2024. Udine, Italy
Degrees of incompleteness of implicative logics: the trichotomy theorem Topology, Algebra, and Categories in Logic (TACL) 2024. Barcelona, Spain
Degrees of incompleteness of implicative logics: the trichotomy theorem PhDs in Logic 2023. Granada, Spain
Degrees of incompleteness of implicative logics: the trichotomy theorem Logic, Algebras and Truth Degrees (LATD) 2023. Tbilisi, Georgia
Sahlqvist theory for fragments of intuitionistic logic Logic, Algebras and Truth Degrees (LATD) 2022. Paestum, Italy
Intuitionistic Sahlqvist theory for deductive systems Topology, Algebra, and Categories in Logic (TACL) 2022. Coimbra, Portugal
Representable forests and diamond systems Boolean Algebras, Lattices, General Algebra, Set Theory, Topology (BLAST) 2021. Las Cruces, NM, USA (online)