FLUX: Geometry-Aware Longitudinal Flow Matching with Mixture of Experts

• URL: http://arxiv.org/abs/2605.08648v1
• Date: Sat, 09 May 2026 03:36:00 GMT
• Title: FLUX: Geometry-Aware Longitudinal Flow Matching with Mixture of Experts
• Abstract summary: We introduce FLUX, a geometry-aware longitudinal flow-matching framework for joint transport modeling and unsupervised regime discovery.<n>Across manifold controls, a regime-switching Lorenz system, widefield cortical calcium imaging during associative learning, and embryoid body single-cell differentiation, FLUX reconstructs longitudinal transport while recovering interpretable regime structure.
• Score: 5.287641390746397
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many biological systems evolve through continuous local dynamics while switching between latent regimes defined by learning, stimulus context, internal state, or developmental stage. These processes are often observed only as unpaired longitudinal snapshots: the same cells, neurons, or animals are not tracked as matched trajectories, even though population states are sampled across successive stages. This creates two coupled challenges. First, trajectories must respect curved low-dimensional manifolds embedded in high-dimensional biological measurements. Second, the model must identify when the transport mechanism itself changes. We introduce FLUX (FLow matching for Unpaired longitudinal data with miXture-of-experts), a geometry-aware longitudinal flow-matching framework for joint transport modeling and unsupervised regime discovery. FLUX learns a data-dependent metric from pooled labeled and unlabeled observations, uses that metric to construct geometry-aware conditional paths between adjacent marginals, and decomposes the resulting velocity field into sparse expert vector fields selected by a Straight-Through Gumbel-Softmax router. Across manifold controls, a regime-switching Lorenz system, widefield cortical calcium imaging during associative learning, and embryoid body single-cell differentiation, FLUX reconstructs longitudinal transport while recovering interpretable regime structure. Ablations show that mixture-of-experts routing alone is insufficient: FLUX without geometric learning can fit local transport but fails or weakens regime discovery when regimes are encoded in local dynamics. These results suggest that geometry-aware velocity decomposition provides a general strategy for discovering latent biological state transitions from unpaired longitudinal snapshots.

Related papers

• Fluid-DiT: Graph-Free Diffusion Transformers for Fluid Flow Simulations Learning [50.053186229257555]
  We propose a Graph-Free Diffusion Transformer that replaces graph message passing with attention-based denoising.<n>By leveraging the transformer's global receptive field, Fluid-DiT naturally captures both local flow structures and long-range correlations.<n>It consistently outperforms graph-based diffusion baselines in both sample quality and distributional accuracy.
  arXiv  Detail & Related papers  (2026-08-07T12:28:45Z)
• ProgFormer: Hierarchical Voxel Diffusion Transformer for Longitudinal Brain MRI Prediction [81.07048249760953]
  longitudinal changes are often subtle and confined to specific anatomical regions.<n>Most subject-specific brain structure remains stable over time.<n>ProgFormer is a hierarchical voxel-space Diffusion Transformer for longitudinal brain MRI prediction.
  arXiv  Detail & Related papers  (2026-07-30T00:13:42Z)
• EFLUX: Elastic Multi-Robot Formation Navigation and Adaptation with Agentic LLMs [24.382645393013945]
  Multi-robot teams operating in confined or cluttered environments must adapt both their formation geometry and group topology to navigate through complex obstacles.<n>Existing methods often model these behaviors independently, connect them through handcrafted rules, or lack explicit geometric criteria for determining when each behavior should be invoked.<n>We propose a geometry-grounded LLM agentic framework for automatic and elastic multi-robot formation navigation.
  arXiv  Detail & Related papers  (2026-07-13T18:08:57Z)
• Liouvillian Geometry of Multidimensional Spectra: Pathway Transport and Observational Holonomy in Open Quantum Systems [4.6612718429485165]
  Liouville pathways provide the conceptual foundation for interpreting multidimensional spectroscopies.<n>In open quantum systems, this picture is incomplete.<n>We develop a geometric framework in which pathway transport is governed by a Liouvillian connection.
  arXiv  Detail & Related papers  (2026-06-21T14:40:22Z)
• Geometry-Driven Flow Analysis of Brain Sulcal Pattern [1.5803208833562954]
  Cortical folding is increasingly recognized as a sensitive marker of neurological disease.<n>In juvenile myoclonic epilepsy, cortical abnormalities are often subtle, spatially distributed, and difficult to detect.<n>We introduce a Poisson-equation-based framework that models cortical folding as a geometry-driven flow.
  arXiv  Detail & Related papers  (2026-06-07T01:41:47Z)
• scHelix: Asymmetric Dual-Stream Integration via Explicit Gene-Level Disentanglement [84.88065404629079]
  scHelix is a dataset-adaptive framework that fundamentally changes how features are processed.<n> scHelix utilizes a dual-stream sparse diffusion encoder equipped with stop-gradient graph caching.<n> scHelix outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-05-18T15:55:13Z)
• TimeLesSeg: Unified Contrast-Agnostic Cross-Sectional and Longitudinal MS Lesion Segmentation via a Stochastic Generative Model [0.338424069286569]
  TimeLesSeg is a unified contrast-agnostic framework designed to segment MS lesions regardless of the presence of a temporal dimension in its inputs.<n>Our approach models pathological priors through lesion masks, which are processed together with the current scan.<n>Cross-sectional processing is enabled by exposing the model to training cases where no prior information is available.
  arXiv  Detail & Related papers  (2026-05-08T16:19:34Z)
• MIOFlow 2.0: A unified framework for inferring cellular stochastic dynamics from single cell and spatial transcriptomics data [47.97758608694957]
  We introduce Interpolating Manifold-Transport Flow (MIOFlow) 2.0.<n>This framework learns biologically informed cellular trajectories by integrating manifold learning, optimal transport, and neural differential equations.<n>It is validated on synthetic datasets, embryoid body differentiation, and spatially resolved axolotl brain regeneration.
  arXiv  Detail & Related papers  (2026-03-23T20:49:45Z)
• SOMA: Unifying Parametric Human Body Models [47.381120553445136]
  Parametric human body models are foundational to human reconstruction, animation, and simulation, yet they remain mutually incompatible.<n>We present SOMA, a unified body layer that transforms these heterogeneous representations through three abstraction layers.<n>The entire pipeline is fully differentiable end-to-end and GPU-accelerated via NVIDIA-Warp.
  arXiv  Detail & Related papers  (2026-03-17T17:58:03Z)
• Learning Discrete Successor Transitions in Continuous Attractor Networks: Emergence, Limits, and Topological Constraints [0.0]
  Continuous attractor networks (CANs) represent low-dimensional continuous variables such as head direction, spatial position, and phase.<n>In canonical spatial domains, transitions along the attractor manifold are driven by continuous displacement signals.<n>We present an experimental framework for training CANs to perform successor-like transitions between stable attractor states.
  arXiv  Detail & Related papers  (2026-01-20T10:41:18Z)
• SKANet: A Cognitive Dual-Stream Framework with Adaptive Modality Fusion for Robust Compound GNSS Interference Classification [47.20483076887704]
  Global Navigation Satellite Systems (GNSS) face growing threats from sophisticated jamming interference.<n>We propose a cognitive deep learning framework built upon a dual-stream architecture that integrates Time-Frequency Images (TFIs) and Power Spectral Density (PSD)<n>We show that SKANet achieves an overall accuracy of 96.99%, exhibiting superior robustness for compound jamming classification.
  arXiv  Detail & Related papers  (2026-01-19T07:42:45Z)
• Flow based approach for Dynamic Temporal Causal models with non-Gaussian or Heteroscedastic Noises [37.02662517645979]
  We introduce FANTOM, a unified framework for causal discovery.<n>It handles non-stationary processes along with non-Gaussian and heteroscedastic noises.<n>It simultaneously infers the number of regimes and their corresponding indices and learns each regime's Directed Acyclic Graph.
  arXiv  Detail & Related papers  (2025-06-20T15:12:43Z)
• FLEX: A Backbone for Diffusion-Based Modeling of Spatio-temporal Physical Systems [51.15230303652732]
  FLEX (F Low EXpert) is a backbone architecture for generative modeling of-temporal physical systems.<n>It reduces the variance of the velocity field in the diffusion model, which helps stabilize training.<n>It achieves accurate predictions for super-resolution and forecasting tasks using as few features as two reverse diffusion steps.
  arXiv  Detail & Related papers  (2025-05-23T00:07:59Z)
• Quadrature nonreciprocity: unidirectional bosonic transmission without breaking time-reversal symmetry [0.0]
  We extend the notion of nonreciprocity to unidirectional bosonic transport in systems with a time-reversal symmetric Hamiltonian. In contrast to standard nonreciprocity, this unidirectional transport manifests when the mode quadratures are resolved with respect to an external reference phase. Our work opens up new avenues for signal routing and quantum-limited amplification in bosonic systems.
  arXiv  Detail & Related papers  (2022-07-18T11:37:54Z)
• Manifold Interpolating Optimal-Transport Flows for Trajectory Inference [64.94020639760026]
  We present a method called Manifold Interpolating Optimal-Transport Flow (MIOFlow) MIOFlow learns, continuous population dynamics from static snapshot samples taken at sporadic timepoints. We evaluate our method on simulated data with bifurcations and merges, as well as scRNA-seq data from embryoid body differentiation, and acute myeloid leukemia treatment.
  arXiv  Detail & Related papers  (2022-06-29T22:19:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.