#world models

437 stories · 37 in 7 days · 124 in 30 days
Oct 15
Oct 14
PapersBuildersSelected20:30
Memento 3 Clears ARC-AGI-3 With a Frozen LLM

Learning here happens entirely outside the model weights: the rulebook-to-code loop acts as an external, verifiable memory, which sidesteps catastrophic forgetting and costly retraining. Builders working on agents for long-horizon games or planning tasks can borrow this pattern of retaining updates only when they replay perfectly against observed data. Against typical fine-tuning pipelines, it trades LLM calls for cheaper local execution.

DAIR.AI@dair_aiSource
ModelsBuildersSelected02:09
Dyna Robotics Explains Dyna-2 World Models on Latent Space

Adapting to new tasks with only 45 minutes of demonstrations suggests world-action models are cutting robot training costs by orders of magnitude. Robotics engineers and automation operators can study how Dyna pairs prediction with action planning and scales via human video. The Din Tai Fung restaurant deployment is a concrete example of physical AI moving past demos into commercial service work.

Latent.Space@latentspacepodSource
Oct 13
Oct 12
Oct 11
Oct 9
ModelsBuildersMulti-sourceSelected00:18
Nvidia Long-WAM framework

The framework demonstrates how extending temporal context dramatically improves robotic task success rates, potentially enabling more complex autonomous behaviors. Developers working on robotics and AI agents should evaluate how this approach could enhance their applications, particularly for tasks requiring longer-term planning and memory.

Event
Aran Komatsuzaki (论文推介)@arankomatsuzakiSource
Oct 8
Oct 7
Oct 6
PapersBuildersSelected22:02
AMI presents H-JEPA world model for hierarchical planning

A 55-percentage-point jump on Visual AntMaze with reduced planning compute suggests hierarchical abstraction is becoming a practical lever for planning efficiency, not just a theoretical idea. Researchers in world models and agent-based control should review the architecture details on the project page. The results also give model builders a concrete baseline to compare against JEPA-style approaches.

Aran Komatsuzaki (论文推介)@arankomatsuzakiSource
Oct 5
Oct 4
Oct 1
Sep 30