AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

StreamSpatial: A Benchmark and Framework for Streaming 3D Visual-Spatial Reasoning

Junlin Xie, Keyang Zhong, Quanlong Zheng, Ruifei Zhang, Kuo Wang, Yanhao Zhang, Haonan Lu, Xiang Wan, Guanbin Li

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply Stream3D-Mem memory and latent cognitive-map reasoningQwen2.5-VL-3B · HourVideo Navigationaccuracy: 32.14 -> 37.50Improved
apply Stream3D-MemQwen2.5-VL-3B · HourVideo Perceptionaccuracy: 35.10 -> 41.39Improved
apply Stream3D-MemQwen2.5-VL-3B · HourVideo Reasoningaccuracy: 31.60 -> 35.32Improved
apply Stream3D-MemQwen2.5-VL-3B · HourVideo Summarizationaccuracy: 50.75 -> 47.56Hurt
add spatial-specific criteria to a simple prompt (ablation)MLLMs · StreamSpatialBenchaccuracy: <=4%No clear effect
add Zero-CoT, streaming checkpoint CoT or memory-replay CoT prompting (ablation)MLLMs · StreamSpatialBenchaccuracyNo clear effect
increase temporal span and region span of the question (ablation)MLLMs · StreamSpatialBench historical tasksaccuracyHurt
increase model scale within a family (ablation)Qwen2.5-VL · StreamSpatialBenchavg accuracy: 34.49 -> 38.41Improved
increase model scale within a family (ablation)LLaVA-Video · StreamSpatialBenchavg accuracy: 38.41 -> 44.01Improved
specialize for spatial (Spatial-MLLM) instead of general-purpose model (ablation)7B-8B models · StreamSpatialBenchavg accuracy: 33.28 -> 37.01Hurt
apply Stream3D-Mem to evaluate offlineStream3D-Mem · VSI-Benchaverage accuracy: 41.10 -> 47.30Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy