Expanding Flow Maps can insert tokens or nodes while denoising
KevinKaichuang · x · 2026-07-28
Expanding Flow Maps let flows grow the output size while denoising
The post introduces Expanding Flow Maps (EFMs), a new family of flow-based generative models that can generate outputs with more dimensions than the input.
Key points:
- Existing flow models usually assume fixed input and output dimensions or fixed sequence length.
- EFMs add an expand operator that inserts new coordinates or tokens conditioned on the current state.
- A transport map then denoises and moves the expanded state forward.
- The framework also extends to the discrete simplex, enabling variable-size graph generation and variable-length sequence generation.
- The authors frame this as useful for cases where output size itself should be a learned, controllable variable, with applications in biology and chemistry.
Related event: New Flow Map Framework Enables Dynamic Output Expansion During Inference(3 posts)→
More from Embodied
- iPhone Duo preorders sell out in 3 minutes around NYC, deliveries slip to late November — BenBajarin · 2026-10-16
- UMD's PAIR framework bridges perception and action in VLA models via bridge tokens — Kaixi Feng · 2026-10-16
- CI-MSE offline metric hits -0.87 Spearman correlation with real robot task success vs weak raw MSE — chris_j_paxton · 2026-10-16
- List of 15 best robotics researchers on X, from Jim Fan to Chelsea Finn and Fei-Fei Li — FinanceYF5 · 2026-10-16
- Fake Apple listing for dual-screen 'iPhone Duo' jokes it's the best phone for watching your AI agents — AIFlow_ML · 2026-10-16
- Inside Auki's Hong Kong office: a fake grocery store powering robotics data collection — LoganGrasby · 2026-10-16