PhD student at Stony Brook University. Working on generative models for images.
I am part of the Computer Vision Lab, advised by
Dimitris Samaras.
You can find more details about our work on diffusion models for the digital histopathology and satellite image domains here.
I also interned at Microsoft Research.
My mentor is
Nebojsa Jojic,
with whom we work on interesting diffusion model problems.
Fast constrained sampling in pre-trained diffusion models
Alexandros Graikos,
Nebojsa Jojic,
Dimitris Samaras
Preprint
Using Newton's method we speed up training-free inference with arbitrary constraints up to 30x. This is the algorithm that powers ZoomLDM.
ZoomLDM: Latent Diffusion Model for multi-scale image generation
Srikar Yellapragada*,
Alexandros Graikos*,
Kostas Triaridis,
Prateek Prasanna,
Rajarsi R. Gupta,
Joel Saltz,
Dimitris Samaras
CVPR, 2025
We propose a multi-scale diffusion model to make generation of massive pathology images feasible.
Learned representation-guided diffusion models for large-image generation
Alexandros Graikos*,
Srikar Yellapragada*,
Minh-Quan Le,
Saarthak Kapse,
Prateek Prasanna,
Joel Saltz,
Dimitris Samaras
CVPR, 2024
We use self-supervised models to provide annotations for diffusion models in digital pathology. Exploiting the unique structure of the data, we synthesize large images with only a patch-based model.
∞-Brush: Controllable Large Image Synthesis with Diffusion Models in Infinite Dimensions
Minh-Quan Le*,
Alexandros Graikos*,
Srikar Yellapragada,
Rajarsi Gupta,
Joel Saltz,
Dimitris Samaras
ECCV, 2024
Using ∞-dimensional diffusion models to generate large pathology and satellite images.
Conditional Generation from Unconditional Diffusion Models using Denoiser Representations
Alexandros Graikos,
Srikar Yellapragada,
Dimitris Samaras
BMVC, 2023
We show that we can learn to control generation with as few as 20 examples using the existing denoiser representations.
GFlowNet-EM for learning compositional latent variable models
Edward Hu,
Nikolay Malkin,
Moksh Jain,
Katie Everett,
Alexandros Graikos,
Yoshua Bengio
ICML, 2023
A GFlowNet can be used as the encoder in a discrete autoencoder model. The flexibility of GFlowNets opens up possibilities for training image encoders with compositional latents.
Diffusion models as plug-and-play priors
Alexandros Graikos,
Nikolay Malkin,
Nebojsa Jojic,
Dimitris Samaras
NeurIPS, 2022
Diffusion models can be used in inference tasks without any additional training. You can even solve the traveling salesman problem with a diffusion model (left).
Resolving label uncertainty with implicit generative models
Esther Rolf,
Nikolay Malkin,
Alexandros Graikos,
Ana Jojic,
Caleb Robinson,
Nebojsa Jojic
UAI, 2022
Images with imprecise annotations can be used to define an implicit generative model. An implicit generative model can resolve the uncertainty of the labels, making the annotations precise.
Single-Image Based Food Volume Estimation
Alexandros Graikos,
Vasileios Charisis,
Dimitrios Iakovakis,
Stelios Hadjidimitriou,
Leontios Hadjileontiadis
HCI International, 2020
Training and applying monocular depth estimation models to estimate the food volume from a given RGB image.
Monocular Depth Estimation using Generative Adversarial Networks
Alexandros Graikos,
Anastasios Delopoulos
Undergraduate Thesis, 2018
Adding GAN losses to monocular depth estimation models improves predicted depth accuracy.
Pathology Image Compression with Pre-trained Autoencoders
Srikar Yellapragada,
Alexandros Graikos,
Kostas Triaridis,
Zilinghan Li,
Tarak Nath Nandi,
Ravi K Madduri,
Prateek Prasanna,
Joel Saltz,
Dimitris Samaras
Preprint
ZoomLDM: Latent Diffusion Model for multi-scale image generation
Srikar Yellapragada*,
Alexandros Graikos*,
Kostas Triaridis,
Prateek Prasanna,
Rajarsi R. Gupta,
Joel Saltz,
Dimitris Samaras
CVPR, 2025
Gen-SIS: Generative Self-augmentation Improves Self-supervised Learning
Varun Belagali*,
Srikar Yellapragada*,
Alexandros Graikos,
Saarthak Kapse,
Zilinghan Li,
Tarak Nath Nandi,
Ravi K Madduri,
Prateek Prasanna,
Joel Saltz,
Dimitris Samaras
Preprint
Fast constrained sampling in pre-trained diffusion models
Alexandros Graikos,
Nebojsa Jojic,
Dimitris Samaras
Preprint
∞-Brush: Controllable Large Image Synthesis with Diffusion Models in Infinite Dimensions
Minh-Quan Le*,
Alexandros Graikos*,
Srikar Yellapragada,
Rajarsi Gupta,
Joel Saltz,
Dimitris Samaras
ECCV, 2024
Diffusion-Refined VQA Annotations for Semi-Supervised Gaze Following
Qiaomu Miao,
Alexandros Graikos,
Jingwei Zhang,
Sounak Mondal,
Minh Hoai,
Dimitris Samaras
ECCV, 2024
Learned representation-guided diffusion models for large-image generation
Alexandros Graikos*,
Srikar Yellapragada*,
Minh-Quan Le,
Saarthak Kapse,
Prateek Prasanna,
Joel Saltz,
Dimitris Samaras
CVPR, 2024
PathLDM: Text conditioned Latent Diffusion Model for Histopathology
Srikar Yellapragada*,
Alexandros Graikos*,
Prateek Prasanna,
Tahsin Kurc,
Joel Saltz,
Dimitris Samaras
WACV, 2024
Conditional Generation from Unconditional Diffusion Models using Denoiser Representations
Alexandros Graikos,
Srikar Yellapragada,
Dimitris Samaras
BMVC, 2023
S-volsdf: Sparse multi-view stereo regularization of neural implicit surfaces
Haoyu Wu,
Alexandros Graikos,
Dimitris Samaras
ICCV, 2023
GFlowNet-EM for learning compositional latent variable models
Edward Hu,
Nikolay Malkin,
Moksh Jain,
Katie Everett,
Alexandros Graikos,
Yoshua Bengio
ICML, 2023
Diffusion models as plug-and-play priors
Alexandros Graikos,
Nikolay Malkin,
Nebojsa Jojic,
Dimitris Samaras
NeurIPS, 2022