AI Paper Atlas

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

Endowing GPT-4 with a Humanoid Body: Building the Bridge Between Off-the-Shelf VLMs and the Physical World

Yingzhao Jian, Zhongan Wang, Yi Yang, Hehe Fan

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents · 2 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 image labels for VLM visual identification (ablation)BiBo · Task success, Sitsuccess rate (%): 48.59 -> 95.84Improved
add majority voting across five VLM instances (ablation)BiBo · Task success, Sitsuccess rate (%): 91.13 -> 95.84Improved
add IK post-optimization (ablation)BiBo · Task success, Liftsuccess rate (%): 6.80 -> 65.42Improved
condition future motion on actual executed motion (ablation)BiBo · Task success, Liftsuccess rate (%): 28.34 -> 65.42Improved
condition future motion on previously generated motion (ablation)BiBo · Task success, Composite Hardsuccess rate (%): 23.81 -> 27.78Improved
condition on actual executed motion (ablation)BiBo · HumanML3DFID (Phy.): 1.414 -> 1.883Hurt
use latent diffusion model instead of plain diffusion (ablation)BiBo · HumanML3DFID: 0.238 -> 0.076Improved
add causal attention in VAE (ablation)BiBo · HumanML3DFID: 0.101 -> 0.076Improved
use LDM (ablation)BiBo · Control accuracy MAE headMAE: 0.0705 -> 0.0310Improved
condition on previously generated motion (ablation)BiBo · Motion discontinuityavg joint acceleration: 0.0698 -> 0.0379Improved
condition on actual executed motion (ablation)BiBo · Motion discontinuityavg joint acceleration: 0.0370 -> 0.0379No clear effect
use online VLM planning instead of ground-truth planBiBo · Task successsuccess rate (%): Hard 42.86 -> 27.78; Reach 98.91 -> 99.18Mixed

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