原帖引用 · 原文
该语言的正文暂不可用，当前显示已有版本。
Rohan Paul@rohanpaul_ai
[2/2] Here's PAN 2 simulating robot hands folding a shirt: sleeves in, sides in, fold it up, flip it over.
The LLM works out which fold comes next, and the 29B model shows how the fabric moves. Each fold builds on the one before, so every step has to hold. https://t.co/0ccvthm3c7
MBZUAI's Institute of Foundation Models (IFM) just released PAN 2, an open foundation world model for physical intelligence.
It's built towards simulative reasoning: letting an agent play out what happens before it acts.
It works with two kinds of reasoning, each with its own job:
> Conceptual: an LLM works out what happens, when, and how the camera moves, then writes it down as text.
> Perceptual: a 29B diffusion model, trained from scratch, turns that text into what you see.
Take the LLM out, and multi-step tasks fall apart: 8.7% of steps right, no plan finished. Getting the order of actions right is the LLM's job.
With both, PAN 2 gets 75.1% of steps right and finishes 34.0% of plans in full, the highest among the models IFM tested. Every step has to hold for a plan to work.
More details in the technical blog:
https://t.co/l6xKPmKT9k
[1/1]
在 X 查看这条帖子 ↗
· UTC+8