We propose a new representation for encoding 3D shapes as neural fields. The representation isdesignedtobecompatible withthetransformer architecture and to benefit both shape reconstruction and shape generation. Existing works on neural fields aregrid-based representations withlatents defined onaregular grid.

We consider the online setting, where the input arrives over time, and irrevocable decisions must be made without knowledge of the future. For all these problems, any online algorithm must incur a cost that is approximately log|I| times the optimal cost in the worst-case, where |I| is the length of theinput.

Recent efforts have shown promising results in this direction by making use of depth-wiserecurrentnetworks.

We view outcome homogenization as an important class ofsystemicharms that arise when we study socialsystems, i.e.harmsthatrequire observing howindividuals aretreated bymanydecision-makers.2 In 2,we conceptually motivate outcome homogenization in the context of algorithmic hiring.