AI Paper Atlas

Research map / Video / Video generation

Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding

Xianjin Wu, Dingkang Liang, Tianrui Feng, Kui Xia, Yumeng Zhang, Xiaofan Li, Xiao Tan, Xiang Bai

ECCV 2026 · Video › Video generation · approach: World-model video generation · 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 Wan2.1 video-diffusion features with adaptive gated fusionVideo-3D LLM · ScanReferAcc@0.5: 51.7 -> 56.2Improved
add generative features with gated fusionVideo-3D LLM · SQA3DEM: 58.6 -> 61.3Improved
add generative features with gated fusionVideo-3D LLM · Scan2CapCIDEr@0.5: 83.8 -> 83.2Hurt
add generative features with gated fusionQwen2.5VL-7B · VSI-Benchavg accuracy: 48.9 -> 50.5Improved
add generative features with gated fusionOpenVLA-OFT · LIBEROavg success rate: 97.0 -> 97.3Improved
add generative featuresOpenVLA-OFT · Click Bell (real robot)success rate: 40% -> 55%Improved
use generative features only, dropping semantic features (ablation)VEGA-3D · Scan2CapCIDEr@0.5: 83.2 -> 25.2Hurt
adaptive gated fusion instead of Add (ablation)VEGA-3D · ScanReferAcc@0.25: 61.5 -> 63.2Improved
adaptive gated fusion instead of Add (ablation)VEGA-3D · SQA3DEM: 61.8 vs 61.3No clear effect
token-level gate instead of channel-level gate (ablation)VEGA-3D · ScanReferAcc@0.25: 61.8 -> 63.2Improved
cross-attention fusion instead of gated fusion (ablation)VEGA-3D · Scan2CapCIDEr@0.5: 83.2 -> 48.8Hurt
DiT-based Wan2.1 encoder instead of UNet-based Stable Video Diffusion (ablation)VEGA-3D · ScanReferAcc@0.25: 61.3 -> 63.2Improved

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