ChronoVision: Temporal Reasoning via Latent State Reconstruction Paper • 2608.05631 • Published 3 days ago • 33
WorldCycle: Self-Verifiable Reinforcement Learning for Long-Horizon Video World Models Paper • 2608.04964 • Published 4 days ago • 11
SwanTale: Unified Multi-Speaker Speech and Audio Generation for Instruct and Zero-Shot Tasks Paper • 2608.02023 • Published 6 days ago • 154