papers

Publications (9)

stat.ML2023

Deep anytime-valid hypothesis testing

Teodora Pandeva, Patrick Forré, Aaditya Ramdas +1

We propose a general framework for constructing powerful, sequential hypothesis tests for a large class of nonparametric testing problems. The null hypothesis for these problems is…

stat.ML2026

E-Scores for (In)Correctness Assessment of Generative Model Outputs

Guneet S. Dhillon, Javier González, Teodora Pandeva +1

While generative models, especially large language models (LLMs), are ubiquitous in today's world, principled mechanisms to assess their (in)correctness are limited. Using the conf…

cs.LG2025

Parallel Sampling from Masked Diffusion Models via Conditional Independence Testing

Iskander Azangulov, Teodora Pandeva, Niranjani Prasad +2

Masked diffusion models (MDMs) offer a compelling alternative to autoregressive models (ARMs) for discrete text generation because they enable parallel token sampling, rather than…

stat.ML2024

Robust Multi-view Co-expression Network Inference

Teodora Pandeva, Martijs Jonker, Leendert Hamoen +2

Unraveling the co-expression of genes across studies enhances the understanding of cellular processes. Inferring gene co-expression networks from transcriptome data presents many c…

stat.ME2024

E-Valuating Classifier Two-Sample Tests

Teodora Pandeva, Tim Bakker, Christian A. Naesseth +1

We introduce a powerful deep classifier two-sample test for high-dimensional data based on E-values, called E-value Classifier Two-Sample Test (E-C2ST). Our test combines ideas fro…

cs.LG2023

Multi-View Independent Component Analysis with Shared and Individual Sources

Teodora Pandeva, Patrick Forré

Independent component analysis (ICA) is a blind source separation method for linear disentanglement of independent latent sources from observed data. We investigate the special set…