Desk archive

Tag: reinforcement learning

AI Tools/Jul 26, 2026/4 min read

Kuaishou’s KwaiKAT Team Releases KAT-Coder-V2.5, an Open-Weight Agentic Coding Model Trained on 100,000+ Verifiable Repositories

The KAT-Coder-V2.5 model, now available in an open-weight variant under Apache-2.0, was trained inside real executable repositories using a novel infrastructure pipeline that reduced environment…

AI Tools/Jul 14, 2026/5 min read

Skyfall AI Introduces MORPHEUS for Continual Reinforcement Learning

Skyfall AI has launched MORPHEUS, a novel benchmark for continual reinforcement learning in persistent enterprise simulations. This platform addresses real-world operational challenges by requiring AI…