Lukas Billera
@lukasbillera.bsky.social
Reposted by Lukas Billera
But when we looked, we couldn't find a sufficiently general justification.
So our preprint, driven by @lukasbillera.bsky.social with assists from @hedwignordlinder.bsky.social, formalizes this, and extends it a little in ways that are trickier to heuristically reason about:
arxiv.org/abs/2511.16599
So our preprint, driven by @lukasbillera.bsky.social with assists from @hedwignordlinder.bsky.social, formalizes this, and extends it a little in ways that are trickier to heuristically reason about:
arxiv.org/abs/2511.16599
Time dependent loss reweighting for flow matching and diffusion models is theoretically justified
This brief note clarifies that, in Generator Matching (which subsumes a large family of flow matching and diffusion models over continuous, manifold, and discrete spaces), both the Bregman divergence ...
arxiv.org
November 21, 2025 at 11:09 PM
But when we looked, we couldn't find a sufficiently general justification.
So our preprint, driven by @lukasbillera.bsky.social with assists from @hedwignordlinder.bsky.social, formalizes this, and extends it a little in ways that are trickier to heuristically reason about:
arxiv.org/abs/2511.16599
So our preprint, driven by @lukasbillera.bsky.social with assists from @hedwignordlinder.bsky.social, formalizes this, and extends it a little in ways that are trickier to heuristically reason about:
arxiv.org/abs/2511.16599
Reposted by Lukas Billera
A technical thread on loss scaling in diffusion and flow matching models (related to a new preprint):
Since the dawn of time, people have been messing with (or dropping entirely) these pesky time-dependent loss scaling terms, mostly because the models train better without them.
Since the dawn of time, people have been messing with (or dropping entirely) these pesky time-dependent loss scaling terms, mostly because the models train better without them.
November 21, 2025 at 11:09 PM
A technical thread on loss scaling in diffusion and flow matching models (related to a new preprint):
Since the dawn of time, people have been messing with (or dropping entirely) these pesky time-dependent loss scaling terms, mostly because the models train better without them.
Since the dawn of time, people have been messing with (or dropping entirely) these pesky time-dependent loss scaling terms, mostly because the models train better without them.
Reposted by Lukas Billera
With my wonderful lab, who mostly aren't on here (except @lukasbillera.bsky.social and @antonoresten.bsky.social ?) we've been tinkering in this space since the end of the summer, but we think this is just too cool to sit on any longer.
The manuscript should be up by tomorrow and I'll drop a link.
The manuscript should be up by tomorrow and I'll drop a link.
November 10, 2025 at 9:10 AM
With my wonderful lab, who mostly aren't on here (except @lukasbillera.bsky.social and @antonoresten.bsky.social ?) we've been tinkering in this space since the end of the summer, but we think this is just too cool to sit on any longer.
The manuscript should be up by tomorrow and I'll drop a link.
The manuscript should be up by tomorrow and I'll drop a link.
Reposted by Lukas Billera
We figured out flow matching over states that change dimension. With "Branching Flows", the model decides how big things must be! This works wherever flow matching works, with discrete, continuous, and manifold states. We think this will unlock some genuinely new capabilities.
November 10, 2025 at 9:10 AM
We figured out flow matching over states that change dimension. With "Branching Flows", the model decides how big things must be! This works wherever flow matching works, with discrete, continuous, and manifold states. We think this will unlock some genuinely new capabilities.