Training Agents Inside of Scalable World Models · HackerPapers discussions

Training Agents Inside of Scalable World Models — mentioned by andy12_

Fri, 30 Jan 2026 12:44:18 GMT

How is it not a world model? The latents of the model apparently encode enough information to represent a semi-consistent interactuable world. Seems enough world-modely to me. Besides, we already know that agents can be trained with these world models successfully. See[1]: > By learning behaviors in imagination, Dreamer 4 is the first agent to obtain diamonds in Minecraft purely from offline data, without environment interaction. Our work provides a scalable recipe for imagination training, marking a step towards intelligent agents [1] https://arxiv.org/pdf/2509.24527

Paper and discussion history. A mention is not an endorsement.

Training Agents Inside of Scalable World Models — mentioned by andy12_

Wed, 31 Dec 2025 11:35:48 GMT

Because the general idea here is that image and video models, when scaled way up, can generalize like text models did[1], and eventually be treated as "world models"[2]; models that can accurately model real world processes. These "world models" then could be used to train embodied agents with RL in an scalable way[3]. The video-slop and image-slop generators is just a way to take advantage of the current research in world models to get more out of it. [1] https://arxiv.org/pdf/2509.20328 [2] https://deepmind.google/blog/genie-3-a-new-frontier-for-worl... [3] https://arxiv.org/pdf/2509.24527

Paper and discussion history. A mention is not an endorsement.

Training Agents Inside of Scalable World Models — HN submission

Tue, 30 Sep 2025 22:04:50 GMT

Dreamer 4

3 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.