AI Paper Atlas

Research map / Video / Video-language understanding

Spatial-TTT: Streaming Visual-based Spatial Intelligence with Test-Time Training

FANGFU LIU, Diankun Wu, Jiawei Chi, Yimo Cai, Yi-Hsin Hung, Xumin Yu, Hao Li, Han Hu, Yongming Rao, Yueqi Duan

ECCV 2026 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 12 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
add spatial-predictive 3D conv to TTT layers (ablation)Spatial-TTT · VSI-BenchAvg: 62.1 -> 64.4Improved
add dense scene-description data (ablation)Spatial-TTT · VSI-BenchAvg: 61.3 -> 64.4Improved
interleave self-attention anchor layers (hybrid) with TTT (ablation)Spatial-TTT · VSI-BenchAvg: 53.9 -> 64.4Improved
add spatial-predictive conv: numerical questions (ablation)Spatial-TTT · VSI-Bench numericalMRA: 60.7 -> 64.0Improved
add hybrid architecture: multiple-choice (ablation)Spatial-TTT · VSI-Bench multiple-choiceaccuracy: 52.4 -> 64.8Improved
use TTT backbone instead of quadratic attentionSpatial-TTT-2B vs Qwen3-VL-2B · 1024 framesTFLOPs: 1403.1 -> 799.4Improved
use TTT backbone instead of quadratic attention: memorySpatial-TTT-2B vs Qwen3-VL-2B · 1024 framespeak memory GB: 21.2 -> 11.9Improved
Spatial-TTT vs strongest proprietary on MindCubeSpatial-TTT-2B · MindCube-Tinyaccuracy: 63.9 -> 76.2Improved
streaming TTT vs context-limited models on long video countingSpatial-TTT-2B · VSI-SUPER-Count 120minMRA: 0.0 -> 38.4Improved

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