Abstract: We propose a sampling theory for signals that are supported on either directed or undirected graphs. The theory follows the same paradigm as classical sampling theory. We show that perfect ...

Abstract: Sampling random values from a discrete Gaussian distribution with high precision is a major and computationally-intensive operation of emerging and existing cryptographic standards. FALCON ...

This repository modifies UDLM-based discrete diffusion models to apply Stratified Hazard Sampling (SHS) only during evaluation (gen-PPL). Training code remains unchanged; only the evaluation sampling ...

Discrete diffusion language models (dLLMs) have recently emerged as a promising alternative to traditional autoregressive approaches, offering the flexibility to generate tokens in arbitrary orders ...

Counterfactual explanations are the backbone of our methodology, and they help us formulate how we approach the generation of counterfactuals alongside the ability to compare similar lines of prior ...

In standard autoregressive generation, an LLM predicts the next-token distribution, samples a discrete token, and then discards the distribution, passing only the sampled token as new input. To ...

Ali Hussain has a background that consists of a career in finance with large financial institutions and in journalism covering business. Vikki Velasquez is a researcher and writer who has managed, ...