Mehrdad Farajtabar

mfarajtabar's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

authored a paper 10 months ago

Scaling Smart: Accelerating Large Language Model Pre-training with Small Model Initialization

Paper • 2409.12903 • Published Sep 19, 2024 • 23

authored a paper about 1 year ago

CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data

Paper • 2404.15653 • Published Apr 24, 2024 • 30

authored 8 papers over 1 year ago

Architecture Matters in Continual Learning

Paper • 2202.00275 • Published Feb 1, 2022 • 1

ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models

Paper • 2310.04564 • Published Oct 6, 2023 • 2

CLIP meets Model Zoo Experts: Pseudo-Supervision for Visual Enhancement

Paper • 2310.14108 • Published Oct 21, 2023 • 1

Reinforce Data, Multiply Impact: Improved Model Accuracy and Robustness with Dataset Reinforcement

Paper • 2303.08983 • Published Mar 15, 2023

LLM in a flash: Efficient Large Language Model Inference with Limited Memory

Paper • 2312.11514 • Published Dec 12, 2023 • 257

Weight subcloning: direct initialization of transformers using larger pretrained ones

Paper • 2312.09299 • Published Dec 14, 2023 • 19

SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding

Paper • 2310.15308 • Published Oct 23, 2023 • 23

TiC-CLIP: Continual Training of CLIP Models

Paper • 2310.16226 • Published Oct 24, 2023 • 9