Peishen Yan

Zillionkkk

2 3

·

AI & ML interests

Unified Multimodal Model, Post-training, on-device LLM inference acceleration, and Federated Learning.

Recent Activity

authored a paper 5 days ago

FedMomentum: Preserving LoRA Training Momentum in Federated Fine-Tuning

authored a paper 5 days ago

FedCod: An Efficient Communication Protocol for Cross-Silo Federated Learning with Coding

authored a paper 5 days ago

SkyMask: Attack-agnostic Robust Federated Learning with Fine-grained Learnable Masks

View all activity

Organizations

None yet

authored 4 papers 5 days ago

FedMomentum: Preserving LoRA Training Momentum in Federated Fine-Tuning

Paper • 2603.08014 • Published Mar 9

FedCod: An Efficient Communication Protocol for Cross-Silo Federated Learning with Coding

Paper • 2501.00216 • Published Jul 21

SkyMask: Attack-agnostic Robust Federated Learning with Fine-grained Learnable Masks

Paper • 2312.12484 • Published Jul 18, 2024

SenseNova-U1.5: Towards Native Unified Visual Intelligence

Paper • 2609.11929 • Published 6 days ago • 257

liked 3 models 5 days ago

MiniMaxAI/MiniMax-H3

Image-Text-to-Video • 33B • Updated Aug 13 • 4.91M • • 5.34k

deepseek-ai/DeepSeek-V4.1-Flash

Image-Text-to-Text • 763B • Updated 6 days ago • 326k • • 2.69k

sensenova/SenseNova-U1.5-8B-MoT

Any-to-Any • 18B • Updated about 14 hours ago • 10.1k • 243

upvoted a paper 5 days ago

SenseNova-U1.5: Towards Native Unified Visual Intelligence

Paper • 2609.11929 • Published 6 days ago • 257

upvoted a paper 3 months ago

LatentSkill: From In-Context Textual Skills to In-Weight Latent Skills for LLM Agents

Paper • 2606.06087 • Published Jun 4 • 68