11 papers
The Tamed Subgradient Unadjusted Langevin Algorithm beyond Convexity
Iosif Lytras, Nikolaos Makras, Sotirios Sabanis
We study the problem of sampling from target distributions whose potentials are simultaneously non-smooth, subject to superlinear gradient growth, and non-convex. We introduce the…
When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis
Hoyoung Lee, Suhwan Park, Seunghan Lee +15
Financial decision-makers face more information than they can directly inspect, making context compression necessary. Yet when large language models (LLMs) compress financial sourc…
Flatness-Aware Stochastic Gradient Langevin Dynamics
Stefano Bruno, Youngsik Hwang, Jaehyeon An +2
Flatness of the loss landscape has been widely studied as an important perspective for understanding the behavior and generalization of deep learning algorithms. Motivated by this…
The Performance Of The Unadjusted Langevin Algorithm Without Smoothness Assumptions
Tim Johnston, Iosif Lytras, Nikolaos Makras +1
In this article, we study the problem of sampling from distributions whose densities are not necessarily smooth nor logconcave. We propose a simple Langevin-based algorithm that do…
One More Question is Enough, Expert Question Decomposition (EQD) Model for Domain Quantitative Reasoning
Mengyu Wang, Sotirios Sabanis, Miguel de Carvalho +2
Domain-specific quantitative reasoning remains a major challenge for large language models (LLMs), especially in fields requiring expert knowledge and complex question answering (Q…
Wasserstein Convergence of Score-based Generative Models under Semiconvexity and Discontinuous Gradients
Stefano Bruno, Sotirios Sabanis
Score-based Generative Models (SGMs) approximate a data distribution by perturbing it with Gaussian noise and subsequently denoising it via a learned reverse diffusion process. The…