Thodoris Kouzelis

eqvae. [ICML'25] EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modeling.

ReDi. [NeurIPS'25 Spotlight] Boosting Generative Image Modeling via Joint Image-Feature Synthesis

DreamSound. [ICASSP'24] Investigating Personalization Methods in Text to Music Generation

Reading-Diffusion. A collection of interesting papers on Diffusion Models

DisfluentFA. A Weakly Supervised Forced Alignment for disluent speech

awesome-LoRA. A curated list of Parameter Efficient Fine-tuning papers with a TL;DR

CoReDi. [ArXiv 2026] Coevolving Representations in Joint Image-Feature Diffusion

KaldiLongAligner. Speech to Text Alignment tool implemented with Python and Kaldi

localdiff-demo. A repo containing demo for Enabling Local Editing in Diffusion Models by Joint and Individual Component Analysis

wsac. This reporsitory code form Weakly Supervised Automaed Audio Captioning via Text Only Training

local_pnp. This repo contains experiments for local editing in Diffusion Models