← Back to Videos

SpatialScout Benchmarks Active 3D Exploration for VLM Agents

Loading video

SpatialScout evaluates whether vision-language agents can actively explore and construct object-level 3D maps, spanning 80 scenes and 3,140 target objects under a 100-turn interaction budget.

Embodied AI VLM Active Mapping 3D Scene Understanding Spatial reasoning Benchmarks SpatialScout

Category: perception

Author: @HuggingPapers

Date: 2026-10-22T00:00:00

Duration: 30.02s

Source: https://x.com/HuggingPapers/status/2113302391839760403

EgoXperience: Learning World-Grounded Egocentric Representations from Human Experience

EgoXperience: Learning World-Grounded Egocentric Representations from Human Experience

@liuziwei7 · 43 views · 2026-10-20

Egocentric Vision3D reconstruction

OneCanvas: reproject video frames onto a single panoramic canvas so a stock 2D VLM reasons in 3D — SOTA on SQA3D, VSI-Bench and SPBench at an order of magnitude less compute

OneCanvas: reproject video frames onto a single panoramic canvas so a stock 2D VLM reasons in 3D — SOTA on SQA3D, VSI-Bench and SPBench at an order of magnitude less compute

@unknown · 491 views · 2026-06-19

VLMSpatial reasoning

Grounded API: Robot Perception and SLAM

Grounded API: Robot Perception and SLAM

@GroundedSI · 795 views · 2026-09-16

Robot perceptionSLAM

Jev 1.14 adds image bounding box generation, now served through the TypeLLM API

Jev 1.14 adds image bounding box generation, now served through the TypeLLM API

@TypeLLM · 108 views · 2026-10-15

SpatialLM: Labeled 3D Room Layouts from Monocular Phone Video

SpatialLM: Labeled 3D Room Layouts from Monocular Phone Video

@0x0SojalSec · 586 views · 2026-10-09

Spatial Intelligence3D Scene Understanding

Reka EdgeQ: An On-Device VLM Running Natively on the Snapdragon Hexagon NPU

Reka EdgeQ: An On-Device VLM Running Natively on the Snapdragon Hexagon NPU

@RekaAILabs · 588 views · 2026-09-22

VLMOn-Device Inference

As an Amazon Associate, we earn from qualifying purchases.