AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Model

Fuhao Li, Wenxuan Song, Han Zhao, Jingbo Wang, Pengxiang Ding, Donglin Wang, Long ZENG, Haoang Li

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 122 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 representation alignment (Spatial Forcing)OpenVLA-OFT · LIBERO averagesuccess rate (%): 97.1 -> 98.5Improved
add spatial representation alignment (Spatial Forcing)OpenVLA-OFT · LIBERO-Longsuccess rate (%): 94.5 -> 96.0Improved
align to SigLIP features instead of no alignment (ablation)OpenVLA-OFT · LIBERO average (component analysis, 1 GPU)success rate (%): 92.7 -> 94.0Improved
use VGGT as alignment target instead of SigLIP (ablation)OpenVLA-OFT · LIBERO average (component analysis, 1 GPU)success rate (%): 94.0 -> 96.9Improved
add positional embedding to VGGT target (ablation)OpenVLA-OFT · LIBERO average (component analysis, 1 GPU)success rate (%): 94.7 -> 96.9Improved
align at layer 24 instead of layer 1 (ablation)OpenVLA-OFT · LIBERO average (component analysis, 1 GPU)success rate (%): 94.6 -> 96.9Improved
align at layer 24 instead of layer 32 (ablation)OpenVLA-OFT · LIBERO average (component analysis, 1 GPU)success rate (%): 94.8 -> 96.9Improved
train 150K instead of 2K iterations with alignment (ablation)OpenVLA-OFT · LIBERO average (component analysis, 1 GPU)success rate (%): 72.7 -> 96.9Improved
add alignment to speed up training (ablation)OpenVLA-OFT · LIBERO training efficiencyspeedup: 3.8x more quicklyImproved
use 5% of data with alignment (ablation)OpenVLA-OFT · LIBERO data efficiencysuccess rate (%): 75.8Improved
use full 100% data instead of 1% with alignment (ablation)OpenVLA-OFT · LIBERO average (component analysis, 1 GPU)success rate (%): 42.3 -> 96.9Improved
set alignment loss weight 0.5 instead of 0 (ablation)OpenVLA-OFT · LIBERO alpha sweepsuccess rate (%): 73.2 -> 93.6Improved

Later papers that compared with this method

1 later accepted paper reports Spatial Forcing as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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