MetaPoint: Unlocking Precise Spatial Control in Agentic Visual Generation 事件

MetaPoint: Unlocking Precise Spatial Control in Agentic Visual Generation arXiv:2606.05031v1 Announce Type: new Abstract: Generative visual models fundamentally struggle with precise spatial control. This arises from a core disconnect: models can process textual descriptions of space but cannot directly map numerical coordinates onto the 2D image canvas. We introduce MetaPoint, a method that bridges this gap by representing a continuous 2D coordinate as a single, special token. Crucially, MetaP