Proposes diffeomorphic optimization for manifold-constrained problems in generative models via flow maps, with Lie-group extensions for protein design showing metric improvements.
hub
Advances in neural information processing systems , volume=
31 Pith papers cite this work. Polarity classification is still indexing.
31
Pith papers citing it
hub tools
citation-role summary
background 3
method 1
citation-polarity summary
representative citing papers
AgentODE uses LLMs to discover ODE structures and infer parameter distributions from aggregate data, recovering consistent structures on benchmarks and RDEB clinical data with 231 observations from 46 patients.
McMg is a learned phase-space multi-channel multigrid preconditioner that maps residuals to corrections for heterogeneous Helmholtz equations and shows fewer iterations than classical and neural baselines in tests.
Panel Flow Matching is a generative method to estimate panel densities from longitudinal data with statistical guarantees under irregular sampling, supporting completion, synthetic data, and classification.
PG-DPO is a new variational framework that replaces Bellman recursion with a Pontryagin-guided adjoint-MC projection for RL under non-exponential discounting and shows gains on hyperbolic and survival benchmarks.
VHYDRO is a support-safe variational hybrid filter that jointly recovers continuous latent states, discrete contact modes, and sparse port-Hamiltonian laws per regime while preventing loss of feasible transitions.
A single-network implicit neural optimal transport method that solves the c-transform via proximal fixed-point iteration for stable, non-adversarial training.
ChaosNetBench is a tunable synthetic benchmark for STGNNs on chaotic lattice dynamics that shows graph models outperform non-graph baselines at high local and global chaos.
LTE-ODE repurposes local truncation error as an unsupervised dynamic attention mask that preserves continuous Neural ODE evolution in stable regions while triggering discrete compensation only at anomaly points in large-scale traffic data.
OGPP is a particle flow-matching method using orbit-space canonicalization and geometric paths that achieves lower error and fewer steps than prior approaches on 3D benchmarks.
ARCH is a hierarchical flow-based generative model that enables tractable conditional intensity computation and arbitrary conditioning for spatiotemporal event distributions.
Paper-replication makes coding agents complete SciML paper replications only when every recorded claim has provenance, comparison evidence, and report coverage in a validated workspace.
OM-Path frames DGP inference as learning a deterministic transport map from reference to posterior inducing variables using probability-flow ODE on Doob-bridged SDE with Onsager-Machlup path regularization, yielding statistically significant gains over DBVI on power and protein UCI datasets.
ICDN is a neural network that models log-demand from log-prices so elasticities can be derived exactly by differentiation, showing better out-of-sample performance than log-log benchmarks on beer sales data.
SNMPP builds a product-form neural influence kernel from a signed interaction network over event classes and a delay-aware monotonic temporal network to enable explicit discovery of inter-event relationships alongside strong prediction.
A Bayesian framework models synapses as point sources under diffeomorphic deformation with Poisson noise to simultaneously track locations, intensities, and tissue motion in longitudinal in vivo microscopy.
Geometric Pareto Control embeds Pareto solutions in a Lie group submanifold and navigates via Riemannian gradient flow to achieve 100% feasibility and low suboptimality in control tasks without retraining.
S4 models exhibit stable time-continuity unlike sensitive S6 models, with task continuity predicting performance and enabling temporal subsampling for better efficiency.
Observational and counterfactual distributions are linked by identical support and invariant features, enabling a flow-matching estimator with semiparametric efficiency correction to generate debiased counterfactuals from observations.
Neural Control uses adjoint differentiation of equilibrium conditions to compute trajectory-dependent proxy gradients for history-dependent implicit models in deformable object manipulation.
A deep kernel learning architecture with transformer feature extraction on clinical-BERT embeddings and Gaussian process backend identifies three glaucoma subgroups by decoupling progression trajectories from current visual acuity in multimodal EHR data.
A single-objective rectified flow variant uses neural ODEs trained by regression to monotonically decrease a fixed convex transport cost while preserving marginal distributions.
S5 uses a single MIMO state space model with S4-derived initialization to match S4 efficiency and reach 87.4% average accuracy on the Long Range Arena benchmark.
GMHF aligns generated data distributions with human beliefs about target physics using cNODE and RL to reduce generalization risk, validated on Duffing oscillator and probabilistic models.
citing papers explorer
-
Diffeomorphic Optimization
Proposes diffeomorphic optimization for manifold-constrained problems in generative models via flow maps, with Lie-group extensions for protein design showing metric improvements.
-
LLM-Guided ODE Discovery and Parameter Inference from Small-Cohort Aggregate Data
AgentODE uses LLMs to discover ODE structures and infer parameter distributions from aggregate data, recovering consistent structures on benchmarks and RDEB clinical data with 231 observations from 46 patients.
-
McMg: A Learned Phase-Space Multi-channel Multigrid Preconditioner for Helmholtz Equation
McMg is a learned phase-space multi-channel multigrid preconditioner that maps residuals to corrections for heterogeneous Helmholtz equations and shows fewer iterations than classical and neural baselines in tests.
-
Panel Flow Matching: A Generative Approach to Learning Distributions of Longitudinal Data
Panel Flow Matching is a generative method to estimate panel densities from longitudinal data with statistical guarantees under irregular sampling, supporting completion, synthetic data, and classification.
-
Beyond the Bellman Recursion: A Pontryagin-Guided Framework for Non-Exponential Discounting
PG-DPO is a new variational framework that replaces Bellman recursion with a Pontryagin-guided adjoint-MC projection for RL under non-exponential discounting and shows gains on hyperbolic and survival benchmarks.
-
Support-Safe Variational Hybrid Filtering for Contact-Mode and Sparse-Law Recovery
VHYDRO is a support-safe variational hybrid filter that jointly recovers continuous latent states, discrete contact modes, and sparse port-Hamiltonian laws per regime while preventing loss of feasible transitions.
-
Implicit Neural Optimal Transport via Fixed-Point Optimization
A single-network implicit neural optimal transport method that solves the c-transform via proximal fixed-point iteration for stable, non-adversarial training.
-
ChaosNetBench: Benchmarking Spatio-Temporal Graph Neural Networks on Chaotic Lattice Dynamics
ChaosNetBench is a tunable synthetic benchmark for STGNNs on chaotic lattice dynamics that shows graph models outperform non-graph baselines at high local and global chaos.
-
Local Truncation Error-Guided Neural ODEs for Large Scale Traffic Forecasting
LTE-ODE repurposes local truncation error as an unsupervised dynamic attention mask that preserves continuous Neural ODE evolution in stable regions while triggering discrete compensation only at anomaly points in large-scale traffic data.
-
Generative Modeling with Orbit-Space Particle Flow Matching
OGPP is a particle flow-matching method using orbit-space canonicalization and geometric paths that achieves lower error and fewer steps than prior approaches on 3D benchmarks.
-
Arbitrarily Conditioned Hierarchical Flows for Spatiotemporal Events
ARCH is a hierarchical flow-based generative model that enables tractable conditional intensity computation and arbitrary conditioning for spatiotemporal event distributions.
-
Coding-agents can replicate scientific machine learning papers
Paper-replication makes coding agents complete SciML paper replications only when every recorded claim has provenance, comparison evidence, and report coverage in a validated workspace.
-
Onsager-Machlup Posterior Transport for Deep Gaussian Processes
OM-Path frames DGP inference as learning a deterministic transport map from reference to posterior inducing variables using probability-flow ODE on Doob-bridged SDE with Onsager-Machlup path regularization, yielding statistically significant gains over DBVI on power and protein UCI datasets.
-
Integrable Elasticity via Neural Demand Potentials
ICDN is a neural network that models log-demand from log-prices so elasticities can be derived exactly by differentiation, showing better out-of-sample performance than log-log benchmarks on beer sales data.
-
Structured Neural Marked Point Processes for Interpretable Event Interaction Modeling
SNMPP builds a product-form neural influence kernel from a signed interaction network over event classes and a delay-aware monotonic temporal network to enable explicit discovery of inter-event relationships alongside strong prediction.
-
Bayesian In Vivo Tracking of Synapses using Joint Poisson Deconvolution and Diffeomorphic Registration
A Bayesian framework models synapses as point sources under diffeomorphic deformation with Poisson noise to simultaneously track locations, intensities, and tissue motion in longitudinal in vivo microscopy.
-
Geometric Pareto Control: Riemannian Gradient Flow of Energy Function via Lie Group Homotopy
Geometric Pareto Control embeds Pareto solutions in a Lie group submanifold and navigates via Riemannian gradient flow to achieve 100% feasibility and low suboptimality in control tasks without retraining.
-
Continuity Laws for Sequential Models
S4 models exhibit stable time-continuity unlike sensitive S6 models, with task continuity predicting performance and enabling temporal subsampling for better efficiency.
-
Debiased Counterfactual Generation via Flow Matching from Observations
Observational and counterfactual distributions are linked by identical support and invariant features, enabling a flow-matching estimator with semiparametric efficiency correction to generate debiased counterfactuals from observations.
-
Neural Control: Adjoint Learning Through Equilibrium Constraints
Neural Control uses adjoint differentiation of equilibrium conditions to compute trajectory-dependent proxy gradients for history-dependent implicit models in deformable object manipulation.
-
Deep Kernel Learning for Stratifying Glaucoma Trajectories
A deep kernel learning architecture with transformer feature extraction on clinical-BERT embeddings and Gaussian process backend identifies three glaucoma subgroups by decoupling progression trajectories from current visual acuity in multimodal EHR data.
-
Rectified Flow: A Marginal Preserving Approach to Optimal Transport
A single-objective rectified flow variant uses neural ODEs trained by regression to monotonically decrease a fixed convex transport cost while preserving marginal distributions.
-
Simplified State Space Layers for Sequence Modeling
S5 uses a single MIMO state space model with S4-derived initialization to match S4 efficiency and reach 87.4% average accuracy on the Long Range Arena benchmark.
-
Human-Machine Collaboration on Generative Meta-Learning: Model and Algorithm
GMHF aligns generated data distributions with human beliefs about target physics using cNODE and RL to reduce generalization risk, validated on Duffing oscillator and probabilistic models.
-
Topological Neural Dynamics: A Neuron-wise Framework for Sequence Modeling
Neuron-wise graph-coupled dynamics (TND) beat layer-wise sequence models on Pong behavior cloning, averaging 17.47 consecutive catches—over 3× the strongest baseline.
-
Stable and Near-Reversible Diffusion ODE Solvers for Image Editing
Near-reversible Runge-Kutta ODE solvers combined with vector-field smoothing deliver more stable and higher-fidelity text-guided edits in diffusion models than exactly reversible schemes.
-
RepFlow: Representation Enhanced Flow Matching for Causal Effect Estimation
RepFlow combines representation learning and conditional flow matching to estimate both point and distributional causal effects while mitigating selection bias via entropically regularized Wasserstein distance on normalized latent representations.
-
Bayesian Rain Field Reconstruction using Commercial Microwave Links and Diffusion Model Priors
Bayesian inverse problem with diffusion model priors for CML-based rain field reconstruction outperforms baselines by preserving rainfall statistics better than Gaussian processes.
-
Differentiable Chemistry in PINNs for Solving Parameterized and Stiff Reaction Systems
A differentiable chemistry solver is added to PINNs along with parameterized network architecture and stiffness-tailored residual weighting to solve initial/boundary value problems, inverse parameter identification, and parameterized PDEs for hydrogen combustion.
-
Accurate, Efficient, and Explainable Deep Learning Approaches for Environmental Science Problems
The work introduces WaLeF/FIDLAr for flood forecasting, CoDiCast for probabilistic weather, and Hypercube-RAG for explainable environmental QA, claiming superior accuracy, efficiency, and interpretability over baselines.
-
There Will Be a Scientific Theory of Deep Learning
A mechanics of the learning process is emerging in deep learning theory, characterized by dynamics, coarse statistics, and falsifiable predictions across idealized settings, limits, laws, hyperparameters, and universal behaviors.