Picture for Farzan Farnia

Farzan Farnia

SPARKE: Scalable Prompt-Aware Diversity Guidance in Diffusion Models via RKE Score

Add code
Jun 11, 2025
Viaarxiv icon

Fusing Cross-modal and Uni-modal Representations: A Kronecker Product Approach

Add code
Jun 10, 2025
Viaarxiv icon

Towards an Explainable Comparison and Alignment of Feature Embeddings

Add code
Jun 06, 2025
Viaarxiv icon

PromptWise: Online Learning for Cost-Aware Prompt Assignment in Generative Models

Add code
May 24, 2025
Viaarxiv icon

DiffPattern-Flex: Efficient Layout Pattern Generation via Discrete Diffusion

Add code
May 07, 2025
Viaarxiv icon

pFedFair: Towards Optimal Group Fairness-Accuracy Trade-off in Heterogeneous Federated Learning

Add code
Mar 19, 2025
Viaarxiv icon

Dissecting CLIP: Decomposition with a Schur Complement-based Approach

Add code
Dec 24, 2024
Viaarxiv icon

Be More Diverse than the Most Diverse: Online Selection of Diverse Mixtures of Generative Models

Add code
Dec 23, 2024
Viaarxiv icon

A Super-pixel-based Approach to the Stable Interpretation of Neural Networks

Add code
Dec 19, 2024
Figure 1 for A Super-pixel-based Approach to the Stable Interpretation of Neural Networks
Figure 2 for A Super-pixel-based Approach to the Stable Interpretation of Neural Networks
Figure 3 for A Super-pixel-based Approach to the Stable Interpretation of Neural Networks
Figure 4 for A Super-pixel-based Approach to the Stable Interpretation of Neural Networks
Viaarxiv icon

Conditional Vendi Score: An Information-Theoretic Approach to Diversity Evaluation of Prompt-based Generative Models

Add code
Nov 05, 2024
Figure 1 for Conditional Vendi Score: An Information-Theoretic Approach to Diversity Evaluation of Prompt-based Generative Models
Figure 2 for Conditional Vendi Score: An Information-Theoretic Approach to Diversity Evaluation of Prompt-based Generative Models
Figure 3 for Conditional Vendi Score: An Information-Theoretic Approach to Diversity Evaluation of Prompt-based Generative Models
Figure 4 for Conditional Vendi Score: An Information-Theoretic Approach to Diversity Evaluation of Prompt-based Generative Models
Viaarxiv icon