Recent advances such as self-consistency and test-time reinforcement learning improve the reliability of LLMs without additional supervision, but do we really know why they work?
We dig into the maths behind it
I'll be here for the whole week, happy to meet for tacos 🌮 and discuss about generative models, OT, uncertainty quantification, and some recent work that I've done on test-time scaling for LLMs
At sunny SD for #NeurIPS2025, ready to present "Sampling by averaging: a multiscale approach to score estimation" tomorrow at 4.30pm poster #602. Stop by if you want to chat about diffusion-based sampling and generative models!
At sunny SD for #NeurIPS2025, ready to present "Sampling by averaging: a multiscale approach to score estimation" tomorrow at 4.30pm poster #602. Stop by if you want to chat about diffusion-based sampling and generative models!
It was great to present joint work with Francesca Crucinio and @odakyildiz today at UAI’25 in Rio de Janeiro! In short, we propose algorithms to tackle inference and learning in latent variable models with non-differentiable probability density
🖇️openreview.net/forum?id=rTqyD…)
If you are at #AISTATS25 🌴 come this afternoon to check out our poster and chat about generative models, experimental design and AI4Science!
📎Poster 22: Deep Optimal Sensor Placement for Black Box Stochastic Simulations
Dream team: @tobias_schrdr, Peter and @canned_urn_wad