Cohere Labs world model course session 4: DreamerV1-V4 architectures in one go
Cohere_Labs · x · 2026-10-16
Cohere Labs' open "World Model from Scratch" series holds Session 4 on Oct 15, 9AM PST, covering Chapter 5 on architectures of modern world models — planning to walk through DreamerV1 to V4 in one session, focusing on architectural deltas and engineering motivations.
Topics span state-space, latent dynamics, autoregressive, diffusion, video, and action-conditioned world models plus hybrid tradeoffs. Host Nahid Alam, a CV researcher building a world model for drones, demonstrates a compact transformer video world model inspired by Cosmos (an excavator arm sand-mining scene), downsized from 3B to fit 8GB/16GB consumer GPUs.
More from Research
- WorldMemArena Accepted at NeurIPS: Memory Storage ≠ Memory Use — xwang_lk · 2026-10-16
- Mech Interp Veteran Rethinks Alignment, Misgeneralization, CoT Oversight — akbirthko · 2026-10-16
- WEDGE paper generates training environments conditioned on navigation agent failures — _akhaliq · 2026-10-16
- RExBench: best coding agent completes only 33% of autonomous research extensions — najoungkim · 2026-10-16
- Google's CLASHarness benchmarks multi-agent conflicts in computer use, finds frontier VLMs struggle — google · 2026-10-16
- Tsinghua team proves vanilla flow-matching sampling can't exceed constant-depth expressive power — SQZ · 2026-10-16