AI Paper Atlas

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

WorldAgen: Unified State-Action Prediction with Test-Time World Model Training

Chi Wan, Kangrui Wang, Yuan Si, Pingyue Zhang, Manling Li

AAAI 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 1 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 test-time training of the world model with LoRAWorldAgen · CALVINaverage sequence length: 3.87 -> 3.93Improved
add test-time training of the world model with LoRAWorldAgen · LIBERO-10average success rate (%): 75.5 -> 79.0Improved
add world modeling (image prediction) to the policy (ablation)WorldAgen · CALVINaverage sequence length: 2.96 -> 3.87Improved
add world modeling (image prediction) to the policy (ablation)WorldAgen · LIBEROaverage success rate (%): 46.5% -> 75.5%Improved
vary LoRA rank from 16 to 256 during TTT (ablation)WorldAgen-TTT · CALVINaverage sequence length: 3.928, 3.918, 3.918, 3.930, 3.923No clear effect
increase TTT data from 6 to 204 samples (ablation)WorldAgen-TTT · CALVINaverage sequence length: 3.871 -> 3.928Improved
increase TTT data further from 204 to 340 (ablation)WorldAgen-TTT · CALVINaverage sequence length: 3.928 -> 3.917Hurt
add Gaussian noise (std 0.1) to observations (ablation)WorldAgen-TTT · CALVINaverage sequence length: 3.93 -> 3.90Hurt
add Gaussian noise (std 0.1) to observations (ablation)WorldAgen-TTT · LIBEROaverage success rate (%): 79.0 -> 78.0Hurt
lengthen action chunk from 5 to 9 (ablation)WorldAgen · CALVINaverage sequence length: 3.87 -> 3.16Hurt
lengthen action chunk from 3 to 7 (ablation)WorldAgen · LIBEROaverage success rate (%): 78.0% -> 47.0%Hurt
increase image chunk length from 1 to 5 (ablation)WorldAgen · CALVINaverage sequence length: 3.87 -> 1.78Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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