LocateAnything Explained: Parallel Box Decoding and the Next Generation of Vision-Language Grounding

LocateAnything Explained: Parallel Box Decoding and the Next Generation of Vision-Language Grounding Paper Code Project Demo Modern detection-and-grounding VLMs treat a bounding box as text: each box becomes a short string of coordinate tokens, decoded one at a time, left to right. This means a model predicts box coordinates one token at a time, despite all coordinates belonging to the same geometric object. The approach inherits the limitations of language modeling rather than exploiting the structure of spatial prediction. The usual fix for the latency half is multi-token prediction (MTP): emit several tokens per step and accept some accuracy loss for throughput. LocateAnything introduces Parallel Box Decoding (PBD), which predicts an entire bounding box as a single atomic unit. This simultaneously improves localization quality and decoding speed. The approach Parallel Box Decoding LocateAnything-3B, is a native-resolution VLM built from a Moon-ViT vision encoder and a Qwen2.5…