SweetSystems

SweetSystems

·

AI & ML interests

AI Security (Models Agents, MCP, Apps), Utilizing AI for Security Engineering as well, Model Development, img/vid gen. Audio/Speech.

Recent Activity

upvoted a paper 5 days ago

NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness

upvoted a collection 5 days ago

WTF GENIUS PAPERS

upvoted a paper 5 days ago

The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement

View all activity

Organizations

upvoted a paper 5 days ago

NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness

Paper • 2609.08183 • Published 10 days ago • 420

upvoted a collection 5 days ago

WTF GENIUS PAPERS

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 379 items • Updated about 3 hours ago • 82

upvoted a paper 5 days ago

The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement

Paper • 2609.11873 • Published 8 days ago • 88

upvoted a collection 28 days ago

EXL3 models

73 items • Updated 7 days ago • 79

upvoted a collection about 1 month ago

Krea 2 LoRAs

A collection of LoRAs for Krea 2 Turbo and Krea 2 Raw • 9 items • Updated Jun 23 • 59

upvoted 2 papers over 1 year ago

Absolute Zero: Reinforced Self-play Reasoning with Zero Data

Paper • 2505.03335 • Published May 6, 2025 • 191

Skywork-Reward: Bag of Tricks for Reward Modeling in LLMs

Paper • 2410.18451 • Published Oct 24, 2024 • 21