Hy-Memory-30B-A3B: Attributed Reinforcement Learning for Long-Term Memory Writing Paper • 2610.22196 • Published 30 days ago • 6
RAZOR Collection Training-free MoE expert pruning by functional replaceability. Six backbones at 25% and 50% expert removal, no recovery training. • 13 items • Updated Oct 1 • 2
From Proprietary to Open-Source: Bridging the Distribution Gap via Multi-Agent Protocol Distillation in Agentic Search Paper • 2607.24280 • Published Jul 27 • 84
DecoEvo: Score-Decoupled Co-Evolution of Solver and Rubric-Generator Skills in Text Space Paper • 2607.25675 • Published Jul 28 • 67
Running on Zero Agents 36 Newborn Article Impact Predict 💻 36 Use title and abstract to predict future academic impact