Picture for Praneeth Vepakomma

Praneeth Vepakomma

Convergence of Mean Shift Algorithms for Large Bandwidths and Simultaneous Accurate Clustering

Add code
Jun 24, 2025
Viaarxiv icon

ABBA: Highly Expressive Hadamard Product Adaptation for Large Language Models

Add code
May 20, 2025
Viaarxiv icon

Safety Subspaces are Not Distinct: A Fine-Tuning Case Study

Add code
May 20, 2025
Viaarxiv icon

HSplitLoRA: A Heterogeneous Split Parameter-Efficient Fine-Tuning Framework for Large Language Models

Add code
May 05, 2025
Viaarxiv icon

Fed-SB: A Silver Bullet for Extreme Communication Efficiency and Performance in (Private) Federated LoRA Fine-Tuning

Add code
Feb 21, 2025
Viaarxiv icon

Tackling Feature and Sample Heterogeneity in Decentralized Multi-Task Learning: A Sheaf-Theoretic Approach

Add code
Feb 03, 2025
Figure 1 for Tackling Feature and Sample Heterogeneity in Decentralized Multi-Task Learning: A Sheaf-Theoretic Approach
Figure 2 for Tackling Feature and Sample Heterogeneity in Decentralized Multi-Task Learning: A Sheaf-Theoretic Approach
Figure 3 for Tackling Feature and Sample Heterogeneity in Decentralized Multi-Task Learning: A Sheaf-Theoretic Approach
Figure 4 for Tackling Feature and Sample Heterogeneity in Decentralized Multi-Task Learning: A Sheaf-Theoretic Approach
Viaarxiv icon

Predicting Survival of Hemodialysis Patients using Federated Learning

Add code
Dec 14, 2024
Viaarxiv icon

Initialization using Update Approximation is a Silver Bullet for Extremely Efficient Low-Rank Fine-Tuning

Add code
Nov 29, 2024
Viaarxiv icon

Exact Aggregation for Federated and Efficient Fine-Tuning of Foundation Models

Add code
Oct 12, 2024
Figure 1 for Exact Aggregation for Federated and Efficient Fine-Tuning of Foundation Models
Figure 2 for Exact Aggregation for Federated and Efficient Fine-Tuning of Foundation Models
Figure 3 for Exact Aggregation for Federated and Efficient Fine-Tuning of Foundation Models
Figure 4 for Exact Aggregation for Federated and Efficient Fine-Tuning of Foundation Models
Viaarxiv icon

Privacy-Preserving Split Learning with Vision Transformers using Patch-Wise Random and Noisy CutMix

Add code
Aug 02, 2024
Viaarxiv icon