Beyond Geometric Closure: Measuring and Diagnosing Mechanical Consistency in Visual Dynamics
We introduce a framework for measuring and diagnosing mechanical consistency in visual dynamics.
ProofPaper ↗
Key points
- Visual accuracy leaves open whether predicted motion obeys dynamics.
- An independent SI audit of rigid closed chains separates geometric admissibility, reaction-free actuator effort, and work--energy balance.
- Continuous scoring measures progress against fixed engineering targets, while reference replay quantifies measurement error.
- With three seeds per setting, we evaluate KeyCLD and CON across 343 planar assemblies at 64/256 pixels and V-JEPA + AC and LingBot-VA on 100 shared spatial test assemblies.
Sources (1)
- [1]Beyond Geometric Closure: Measuring and Diagnosing Mechanical Consistency in Visual DynamicsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 17, 10:50 AM
We introduce a framework for measuring and diagnosing mechanical consistency in visual dynamics.
Visual accuracy leaves open whether predicted motion obeys dynamics.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 16, 2026Beyond Action Prediction: Learning Search-Free Planning in Latent Space
- Oct 16, 2026Prune the Leaky Neurons: Mitigating Private Data Leakage in Fine-Tuned LLMs via Joint-Embedding-Guided Pruning
- Oct 15, 2026VeriVLA: Adaptive Test-Time Verification for Flow-Based VLAs with Latent World Models
- Oct 15, 2026TS-WAM: Temporal Latent Actions and World-Action Coupling for Vision-Language Robot Control
- Oct 15, 2026What Should Robot Policies Learn to Predict? Rethinking Supervision Targets in Vision-Language-Action Learning