Fig.1

Concept

Boxes per second

Boxes per second measures detection throughput: how many bounding boxes a model can emit per second of wall-clock time, including whatever text or coordinate generation the box requires. It matters for vision-language grounding models because many of them decode box…

The rest of “Boxes per second” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.

Log in to unlock

← Back to the library

Boxes per second, explained · Fig. 1