jc;ml;sc

Organizers

Zac Cross {zacc}† Mathew Huerta‑Enochian {mhuertae}† Joseph McLaughlin {jmclaug2}†

† @uoregon.edu

2026.04.02 (Week 1)

Paper — Equivariant Graph Neural Operator for Modeling 3D Dynamics
Presenter — Eric Zander

Upcoming

2026.04.09 (Week 2)

Paper —
Presenter — Joseph McLaughlin

2026.04.16 (Week 3)

Paper —
Presenter — Simon Casey

2026.04.23 (Week 4)

Paper —
Presenter — Manish Mathai

2026.04.30 (Week 5)

Paper —
Presenter — Mathew Huerta-Enochian

2026.05.07 (Week 6)

Paper —
Presenter — Zacc Cross

Past meetings

2026.03.12 (Week 10w26) KAN-ODEs

Paper — KAN-ODEs: Kolmogorov-Arnold Network Ordinary Differential Equations for Learning Dynamical Systems and Hidden Physics
Presenter — Said Efendiyev
« Link to talk »
In attendance — Simon Casey; Zac Cross; Michael Dushkoff; Said Efendiyev; Mathew Huerta-Enochian; Manish Mathai; Joseph McLaughlin; Eric Zander.
Cookies — provided by Michael.

2026.03.05 (Week 9w26) RenderFlow

Paper — RenderFlow: Single-Step Neural Rendering via Flow Matching
Presenter — Manish Mathai
« Link to talk »
Pathtracing demo — Link
In attendance — Simon Casey; Zac Cross; Said Efendiyev; Kai Iverson; Mathew Huerta-Enochian; Manish Mathai; Joseph McLaughlin; Eric Zander; Yiyang Zhang.
Cookies — provided by Eric.
Remarks — FM has fewers steps during inference and makes it a great candidate for speeding up image processing applications, and the authors claim it these fewer steps introduce less "fuzziness" and noise. Maybe! It seems like whether you choose Diff or FM you're basically working with the same roll of tools, at least to many of us.

2026.02.26 (Week 8w26) FLOWER

Paper — FLOWER: A Flow-Matching Solver for Inverse Problems
Presenter — Zac Cross
« Link to talk »
In attendance — Simon Casey; Zac Cross; Said Efendiyev; Kai Iverson; Mathew Huerta-Enochian; Manish Mathai; Joseph McLaughlin; Tamanna Saini; Eric Zander.

2026.02.20 (Week 7w26) FM Primer

Paper — Flow Matching Primer (a) (b) (c)
Presenter — Mathew Huerta-Enochian
Missing... (pester Mathew)
About — These three papers form much of the foundational theory of FM. If you can read all or some of them, please do so! If your time is constrained, we recommend reading them in the listed order.
In attendance — Zac Cross; Michael Dushkoff; Said Efendiyev; Mathew Huerta-Enochian; Manish Mathai; Joseph McLaughlin; Tamanna Saini; Eric Zander; Yiyang Zhang.
Remarks — MH: Standard manifold hypothesis is key assumption of FM, does this actually hold? MM: How did the authors generate the same images in Lipman? Overfitting on a small dataset? (b) does a diffusion-like operation to straighten the flows to more closely approximate OT (note that we are still working within the SMH here). The rationale for straighter flows is faster generation. (c) OT mini batches to learn OT over the whole model. MH: Octagauss? OT fields can cross as long as they do not cross at the same time. MD: why doesn't (c) do a quantitive analysis on the manifold? TS: Greedy vs stochastic OT. MD: Are there limitations when emerge when the problem is high dimensional? (Maybe). EZ: Non-gaussian sources? Varying the gaussian has an impact. Could you do transfer learning this way? TS: applications of greedy OT vs stochastic. MM: How robust is FM to changes in the source? Example application - denoising. JM: does the SMH make sense? Even if the objective is lower-dimension, is it strictly continuous? MH: This is why CNFs are cool(er than FM) because they don't assume the SMH.

2026.02.13 (Week 6w26) DiffusionSat

Paper — DiffusionSat (2024)
Presenter — Michael Dushkoff
« Link to talk »
In attendance — Zac Cross; Michael Dushkoff; Said Efendiyev; Mathew Huerta-Enochian; Hritvik Jekki Venkateshwarulu; Manish Mathai; Joseph McLaughlin; Eric Zander.
Comments — Stable diffusion trained on clip with a control net. Would this generalize or is this paper just an impressive demonstration? Interesting enough to try in Michael's applied microscope tachography research.

More papers from an even less organized time.