Two auxiliary environments suffice to identify causal graphs and functional mechanisms in structural causal models under acyclicity and invariance assumptions, enabling correct counterfactual inference.
Estimation of Non-Normalized Statistical Models by Score Matching
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
years
2026 7representative citing papers
LHSD uses spectral filtering on the log-density Hessian to isolate tangent directions from noise and estimate local intrinsic dimension scalably via Stochastic Lanczos Quadrature.
TriOpt recovers topological ordering via Sherman-Morrison rank-1 updates on linear kernels and then solves a convex continuous optimization for the linear DAG structure.
A learnable continuous perturbation framework for LLM token prefixes via latent vector transformations, optimized through unbiased estimating equations, yields gains in out-of-domain performance.
Training and sampling in static scalar energy generative models are two instances of the same Lyapunov-driven density transport dynamics on Wasserstein space, differing only by initial condition, which yields a finite stopping criterion for Langevin sampling and additive composition rules that keep
A unified framework for exponential tilting in diffusion and flow models that includes bias-variance decompositions showing finite gradient variance for some methods, norm bounds on adjoint ODEs, and adapted losses with new Crooks and Jarzynski identities.
citing papers explorer
-
Causal Learning with the Invariance Principle
Two auxiliary environments suffice to identify causal graphs and functional mechanisms in structural causal models under acyclicity and invariance assumptions, enabling correct counterfactual inference.
-
Local Hessian Spectral Filtering for Robust Intrinsic Dimension Estimation
LHSD uses spectral filtering on the log-density Hessian to isolate tangent directions from noise and estimate local intrinsic dimension scalably via Stochastic Lanczos Quadrature.
-
TriOpt: A Scalable Algorithm for Linear Causal Discovery
TriOpt recovers topological ordering via Sherman-Morrison rank-1 updates on linear kernels and then solves a convex continuous optimization for the linear DAG structure.
-
Learning Perturbations to Extrapolate Your LLM
A learnable continuous perturbation framework for LLM token prefixes via latent vector transformations, optimized through unbiased estimating equations, yields gains in out-of-domain performance.
-
Energy Generative Modeling: A Lyapunov-based Energy Matching Perspective
Training and sampling in static scalar energy generative models are two instances of the same Lyapunov-driven density transport dynamics on Wasserstein space, differing only by initial condition, which yields a finite stopping criterion for Langevin sampling and additive composition rules that keep
-
A unified perspective on fine-tuning and sampling with diffusion and flow models
A unified framework for exponential tilting in diffusion and flow models that includes bias-variance decompositions showing finite gradient variance for some methods, norm bounds on adjoint ODEs, and adapted losses with new Crooks and Jarzynski identities.
- Reducing Diffusion Model Memorization with Higher Order Langevin Dynamics