MIST Framework Separates Execution from Memory to Diagnose Video World Model Failures
hkuhk · hf · 2026-10-19
Researchers from the University of Hong Kong, University of Michigan, and Peking University propose MIST, a framework that decouples memory tests for video world models into scene setup and change execution, pinpointing whether low scores stem from forgetting or failed execution — current benchmarks fold both into a single number. Building on this diagnosis, their M3-Agent improves instruction execution and significantly boosts benchmark scores. The paper appeared on arXiv on Oct 16.
More from Models
- Together AI teases mysterious 'Space Bunny' launch — togethercompute · 2026-10-19
- Pangram AI detector flags The Atlantic's human-written cover story as 100% AI — nitarshan · 2026-10-19
- Can a world model run in real time on just 12GB of VRAM? — yehiaserag · 2026-10-19
- Memory-native MoE MemT 2.0 (293B) keeps memory inside the forward pass, no retraining needed — MaterialButterfly135 · 2026-10-19
- Reddit test finds Nemotron 3.5 Super VL best at obscure knowledge, beating GPT-OSS and Qwen — Kahvana · 2026-10-19
- Thought experiment: if billion-dollar companies stopped open-weight releases for a year — ForsookComparison · 2026-10-19