Farzan Farnia

DAK-UCB: Diversity-Aware Prompt Routing for LLMs and Generative Models

BALROG: Contextual Bandits meets Active Learning for Online Generative Model Selection

Training-Free Distribution Adaptation for Diffusion Models via Maximum Mean Discrepancy Guidance

Consistency Flow Model Achieves One-step Denoising Error Correction Codes

Towards an Explainable Comparison and Alignment of Feature Embeddings

miniF2F-Lean Revisited: Reviewing Limitations and Charting a Path Forward

When Kernels Multiply, Clusters Unify: Fusing Embeddings with the Kronecker Product

Multilayer Matrix Factorization via Dimension-Reducing Diffusion Variational Inference

Boosting the visual interpretability of CLIP via adversarial fine-tuning

APOLLO: Automated LLM and Lean Collaboration for Advanced Formal Reasoning

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

PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models

Be More Diverse than the Most Diverse: Optimal Mixtures of Generative Models via Mixture-UCB Bandit Algorithms

Certifiably Robust Model Evaluation in Federated Learning under Meta-Distributional Shifts

PAK-UCB Contextual Bandit: An Online Learning Approach to Prompt-Aware Selection of Generative Models and LLMs

Kernel-based Unsupervised Embedding Alignment for Enhanced Visual Representation in Vision-language Models

An Online Learning Approach to Prompt-based Selection of Generative Models

MoreauPruner: Robust Structured Pruning of Large Language Models against Weight Perturbations

Robust Model Evaluation over Large-scale Federated Networks

On the Mode-Seeking Properties of Langevin Dynamics

An Information-Theoretic Approach to Diversity Evaluation of Prompt-based Generative Models