AI & ML
Verify2Act: Critic-Guided Latent World Models for Verifying Language-Conditioned Manipulation Plans
arXiv cs.RO0 views
arXiv:2611.25243v1 Announce Type: new Abstract: Vision-Language Models (VLMs) can propose multi-step manipulation plans, but they have no grounded model of how a scene will change, and the pixel-level world models that could check those plans are slow and drift over long horizons. We present Verify2Act, a modular verification framework that separates semantic action proposal from latent physical verification. A VLM proposes candidate plans, a latent world model (V2A-WM) imagines each of them in the DINOv2 feature space, and a dual-head critic checks the imagined rollouts for temporal consistency and goal satisfaction before anything is executed. V2A-WM extends recent flow-matching world models with cross-attention action grounding and a causal temporal history window. We deploy the full system on a low-cost robot arm, training the world model and critic only on data generated in a digital twin. Across six language-conditioned tasks, Verify2Act reaches 83.3\% task success, compared with 61.7\% for a latent world model without our architectural changes, 53.3\% for the VLM alone, and 46.7\% for a pixel-diffusion world model. It verified every plan it executed and incurred no planning-induced failures: all of its failed episodes were caused by low-level grasp or placement slips. In simulation, Verify2Act obtains the highest five-task chain success on CALVIN among the compared baselines with 30$\times$ fewer reflective VLM calls than pixel-diffusion verification, and the highest success on cluttered nut assembly.
Source:https://arxiv.org/abs/2611.25243PreviousFerrofluid Shape Control with Deep Reinforcement LearningNextA Synthetic Multi-Modal Testbed for Security Stress Testing of Agentic Counter-UAS Systems
Related posts
내 PDF 문서를 로컬 LLM에게 검색시키기 — RAG 실전 구축
0dev.to
How we tell parents what a teen's AI chat was about, without showing them the chat
0dev.to
GitHub Actions GITHUB_TOKEN: 11 Tags Pushed, 0 Release Workflows Ran
0dev.to

I can't believe the podcast I learned to code with is back!
0dev.to
Practice unaided and AI-assisted coding as two different skills
0dev.to
Best Semrush AI Visibility Toolkit Alternatives in 2026 (prices checked)
0dev.to