9 papers
A zero-one law for one-shot system identification
Nicolas Boullé, Diana Halikias, Samuel E. Otto +1
Can a model be identified from one experiment? We study analytic systems that are linearly parameterized by a combination of prescribed dictionary terms, such as partial differenti…
Jacobian Scopes: token-level causal attributions in LLMs
Toni J. B. Liu, Baran ZadeoÄlu, Nicolas Boullé +3
Large language models (LLMs) make next-token predictions based on clues present in their context, such as semantic descriptions and in-context examples. Yet, elucidating which prio…
Physics-guided correction for operator learning under model misspecification
Lei Ma, Nicolas Boullé, Yu-Sen Yang +2
Physics-informed operator learning provides an efficient framework for approximating solution operators of partial differential equations by combining observational data with gover…
Generalized Discrete Diffusion from Snapshots
Oussama Zekri, Théo Uscidda, Nicolas Boullé +1
We introduce Generalized Discrete Diffusion from Snapshots (GDDS), a unified framework for discrete diffusion modeling that supports arbitrary noising processes over large discrete…
Trustworthy Koopman Operator Learning: Invariance Diagnostics and Error Bounds
Gustav Conradie, Nicolas Boullé, Jean-Christophe Loiseau +2
Koopman operator theory provides a global linear representation of nonlinear dynamics and underpins many data-driven methods. In practice, however, finite-dimensional feature space…
Text-Trained LLMs Can Zero-Shot Extrapolate PDE Dynamics, Revealing a Three-Stage In-Context Learning Mechanism
Jiajun Bao, Nicolas Boullé, Toni J. B. Liu +2
Large language models (LLMs) have demonstrated emergent in-context learning (ICL) capabilities across a range of tasks, including zero-shot time-series forecasting. We show that te…