Research

The paper develops a mathematical framework explaining how diffusion models learn low-dimensional data distributions from a finite number of training samples without the curse of dimensionality. It models data as a mixture of low-rank Gaussians and shows that the diffusion training objective is equivalent to subspace clustering over the training samples, so sample complexity scales linearly with intrinsic dimension rather than exponentially with ambient dimension. Experiments on synthetic and real image datasets show phase transitions in generalization.

This paper shows that full likelihood-based inference is feasible for exponential family graphical models, including partially observed ones with latent nodes, under a logarithmically sparse setting. The method uses the tractable independence model underlying an intractable graphical model to estimate the normalizing constant and its gradient. Numerical results, theory, and comparisons with pseudo likelihood-based approaches support its applicability.

COAT (Causal representation Assistant) is a framework that uses large language models to propose measured variables for causal discovery from unstructured data. Rather than inferring causality directly, it feeds intermediate causal discovery results back to the LLM to refine those variables. The authors report theoretical guarantees and evaluate the approach on 7 benchmarks and real-world case studies, including analysis of human reviews and diagnosis of neuropathic and brain tumors.

This paper studies smooth bilevel optimization with a strongly convex lower-level problem and a possibly nonconvex upper-level problem, under stochastic gradients with heavy-tailed noise. The authors propose a nested-loop method, N$^2$SBA, reaching an $\epsilon$-stationary point with SFO complexity $\tilde{\mathcal{O}}\big(\kappa^{\frac{7p-3}{p-1}} \sigma^{\frac{p}{p-1}} \epsilon^{-\frac{4 p - 2}{p-1}}\big)$, where $p\in(1,2]$ is the noise moment order. For nonconvex-strongly-concave minimax problems the method achieves a matching-style bound, and all bounds match the best known results at $p=2$.

The paper presents a method for jointly identifying the drift and diffusion of a stochastic differential equation (SDE) from its temporal marginals alone, without observing individual trajectories. Under linear drift and additive diffusion, non-identifiability arises only if the initial distribution has generalized rotational symmetries, and the authors prove that drift and diffusion can almost always be recovered otherwise. They also show the causal graph can be recovered from the identified parameters, and validate their iterative algorithm APPEX on simulated linear additive noise SDE data.

The paper examines whether a sparse set of symbolic interactions can faithfully explain the inference patterns of large language models, and uses interactions to diagnose LLM representations. The authors find that 50 to 100 interactions are usually enough to mimic changes in an LLM's output scores across an exponential number of masked inputs. They also report that simple interactions are more consistently extracted across architectures and semantically similar inputs than complex ones, and that low-order interactions can guide model compression.

Meta-ARVDM is a unified analytical framework that studies history forgetting and temporal degradation in Auto-Regressive Video Diffusion Models (AR-VDMs). The authors show that history forgetting is characterized by the conditional mutual information between the generated output and preceding frames, and prove that adding more past frames monotonically alleviates it. They also show that temporal degradation can be quantified by the cumulative sum of per-step errors, and report a strong empirical correlation between the two phenomena.

This paper sets out a mathematical foundation for denoising Markov models, a class of generative models pairing a forward process from the target distribution to a simple distribution with a constructed backward process for sampling. Using connections to nonequilibrium statistical mechanics and generalized Doob's h-transform, the authors give a unified variational objective and a recipe for designing such models driven by arbitrary Lévy-type processes. They demonstrate novel models using geometric Brownian motion and jump processes as forward dynamics.

This paper develops a likelihood-based method to estimate edge weights from observed information diffusion paths under a proposed General Linear Threshold (GLT) model, a discrete-time class that includes the linear threshold (LT) and independent cascade (IC) models. The authors derive a finite-sample error bound and show the estimator is asymptotically normal under mild conditions. Experiments on synthetic and real-world networks report improved estimation of spread, node activation prediction, and Influence Maximization solution quality.

This study examines whether borrowers' call activity and online social activity predict default on micro-lending platforms, drawing on social capital theory's bonding and bridging capital. Using 154,579 loan records from 10 Indonesian micro-lending platforms and multiple robustness tests, the authors find that call activity frequency and stability significantly reduce default risk. Online social activity frequency and stability, by contrast, significantly increase it.

The paper proposes PrEPoA, a fine-tuning paradigm for source-free domain adaptation on time-series data that combines prior evaluation of model salience with posterior adaptation. Model salience is treated as a quantifiable proxy of semantic interpretability, measured by a key pattern reconstruction module and calibrated with an interpattern triplet loss. The authors report that PrEPoA outperforms nine UDA and seven SFDA methods across five datasets and works as a plug-and-play module for other SFDA methods.

This paper introduces neural machine unranking (NuMuR), a task of removing selected data from neural information retrieval systems, motivated by data privacy compliance. The authors propose CoCoL, a dual-objective framework combining a contrastive loss that lowers relevance scores on forget sets with a consistent loss that preserves accuracy on retain sets. Experiments on two datasets and four neural IR models show substantial forgetting with minimal retention and generalization loss.

This review examines cybersecurity threats facing elderly people as digitalization expands their access to communication, information and social participation. It covers common cybercrimes, the vulnerabilities that make older users attractive targets, and existing approaches to reducing those risks, and it identifies gaps in current research.