Oguzhan Ersoy

oguzer's activity

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

authored 2 papers about 1 hour ago

Training-Free Dynamic Upcycling of Expert Language Models

Paper • 2603.29765 • Published 13 days ago • 9

Backdoor Attacks on Decentralised Post-Training

Paper • 2604.02372 • Published 14 days ago • 10

upvoted a paper about 9 hours ago

Backdoor Attacks on Decentralised Post-Training

Paper • 2604.02372 • Published 14 days ago • 10

submitted a paper to Daily Papers about 9 hours ago

Backdoor Attacks on Decentralised Post-Training

Paper • 2604.02372 • Published 14 days ago • 10

upvoted a paper about 9 hours ago

Training-Free Dynamic Upcycling of Expert Language Models

Paper • 2603.29765 • Published 13 days ago • 9

authored a paper 5 months ago

Hail to the Thief: Exploring Attacks and Defenses in Decentralised GRPO

Paper • 2511.09780 • Published Nov 12, 2025 • 29

updated a collection 5 months ago

Gensyn Papers

Collection

8 items • Updated 12 days ago • 8

upvoted a paper 5 months ago

Hail to the Thief: Exploring Attacks and Defenses in Decentralised GRPO

Paper • 2511.09780 • Published Nov 12, 2025 • 29

commented a paper 5 months ago

Hail to the Thief: Exploring Attacks and Defenses in Decentralised GRPO

Paper • 2511.09780 • Published Nov 12, 2025 • 29 •

4

upvoted a paper 7 months ago

Sharing is Caring: Efficient LM Post-Training with Collective RL Experience Sharing

Paper • 2509.08721 • Published Sep 10, 2025 • 665

authored a paper 10 months ago

All is Not Lost: LLM Recovery without Checkpoints

Paper • 2506.15461 • Published Jun 18, 2025 • 40

upvoted 5 papers 10 months ago

HDEE: Heterogeneous Domain Expert Ensemble

Paper • 2502.19385 • Published Feb 26, 2025 • 5

Verde: Verification via Refereed Delegation for Machine Learning Programs

Paper • 2502.19405 • Published Feb 26, 2025 • 9

SkipPipe: Partial and Reordered Pipelining Framework for Training LLMs in Heterogeneous Networks

Paper • 2502.19913 • Published Feb 27, 2025 • 6

NoLoCo: No-all-reduce Low Communication Training Method for Large Models

Paper • 2506.10911 • Published Jun 12, 2025 • 9

All is Not Lost: LLM Recovery without Checkpoints

Paper • 2506.15461 • Published Jun 18, 2025 • 40

authored 4 papers 10 months ago

NoLoCo: No-all-reduce Low Communication Training Method for Large Models

Paper • 2506.10911 • Published Jun 12, 2025 • 9

Verde: Verification via Refereed Delegation for Machine Learning Programs

Paper • 2502.19405 • Published Feb 26, 2025 • 9

SkipPipe: Partial and Reordered Pipelining Framework for Training LLMs in Heterogeneous Networks

Paper • 2502.19913 • Published Feb 27, 2025 • 6

HDEE: Heterogeneous Domain Expert Ensemble

Paper • 2502.19385 • Published Feb 26, 2025 • 5