21 46 27

Pavlo Molchanov PRO

pmolchanov

https://www.pmolchanov.com

AI & ML interests

Efficiency in Multi-Modal LLMs

Recent Activity

authored a paper 3 days ago

PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation

authored a paper 3 days ago

EoRA: Training-free Compensation for Compressed LLM with Eigenspace Low-Rank Approximation

authored a paper 3 days ago

Hymba: A Hybrid-head Architecture for Small Language Models

View all activity

Organizations

pmolchanov's activity

upvoted a paper 5 days ago

CLIMB: CLustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training

Paper • 2504.13161 • Published 6 days ago • 86

upvoted 2 papers 7 days ago

Efficient Hybrid Language Model Compression through Group-Aware SSM Pruning

Paper • 2504.11409 • Published 8 days ago • 10

Efficient Reasoning Models: A Survey

Paper • 2504.10903 • Published 9 days ago • 18

upvoted a paper 9 days ago

Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models

Paper • 2504.03624 • Published 19 days ago • 13

upvoted a paper 15 days ago

One-Minute Video Generation with Test-Time Training

Paper • 2504.05298 • Published 16 days ago • 98

upvoted a paper 29 days ago

Scaling Vision Pre-Training to 4K Resolution

Paper • 2503.19903 • Published 29 days ago • 40

upvoted 2 articles 3 months ago

Article

Bamba: Inference-Efficient Hybrid Mamba2 Model

Dec 18, 2024

• 51

Article

Finally, a Replacement for BERT: Introducing ModernBERT

Dec 19, 2024

• 610

upvoted a paper 3 months ago

SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer

Paper • 2501.18427 • Published Jan 30 • 19

upvoted a collection 4 months ago

Cosmos

Collection

The collection of Cosmos models • 31 items • Updated about 12 hours ago • 283

upvoted 2 papers 5 months ago

NVILA: Efficient Frontier Visual Language Models

Paper • 2412.04468 • Published Dec 5, 2024 • 60

Cautious Optimizers: Improving Training with One Line of Code

Paper • 2411.16085 • Published Nov 25, 2024 • 21

upvoted a collection 5 months ago

Hymba

Collection

A series of Hybrid Small Language Models. • 2 items • Updated about 12 hours ago • 29

upvoted 2 papers 5 months ago

Star Attention: Efficient LLM Inference over Long Sequences

Paper • 2411.17116 • Published Nov 26, 2024 • 55

Hymba: A Hybrid-head Architecture for Small Language Models

Paper • 2411.13676 • Published Nov 20, 2024 • 45

upvoted 2 papers 6 months ago

ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

Paper • 2410.21465 • Published Oct 28, 2024 • 11

COAT: Compressing Optimizer states and Activation for Memory-Efficient FP8 Training

Paper • 2410.19313 • Published Oct 25, 2024 • 19

upvoted 3 papers 7 months ago

PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation

Paper • 2410.01680 • Published Oct 2, 2024 • 36

Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction

Paper • 2409.18124 • Published Sep 26, 2024 • 34

MaskLLM: Learnable Semi-Structured Sparsity for Large Language Models

Paper • 2409.17481 • Published Sep 26, 2024 • 48