Knowledge Graph
Graph explorer
Search resolved entities, walk their edges, and read the sentence behind every one.
Entities25 of 207,987
- large language modeltechnology1,043 mentions
- robotic manipulationmarket824 mentions
- artificial intelligencetechnology739 mentions
- reinforcement learningtechnology491 mentions
- density-functional theorytechnology386 mentions
- Transformertechnology371 mentions
- agentic AItechnology353 mentions
- vision-language-action modeltechnology350 mentions
- U.S. Food and Drug Administrationorganization320 mentions
- Vision-Language Modeltechnology318 mentions
- Microsoftorganization291 mentions
- Googleorganization276 mentions
- NVIDIAorganization264 mentions
- Model Context Protocoltechnology232 mentions
- machine learningtechnology204 mentions
- MarketsandMarketsorganization199 mentions
- Claudeproduct192 mentions
- AI agenttechnology188 mentions
- ChatGPTproduct178 mentions
- Anthropicorganization177 mentions
- integrated sensing and communicationtechnology175 mentions
- Amazon Web Services, Inc.organization174 mentions
- OpenAIorganization172 mentions
- diffusion modeltechnology169 mentions
- IBMorganization163 mentions
Entity
world-action model
technology122 mentionslast seen Nov 12, 2026
A model trained on action-labeled robot demonstrations that jointly denoises latent actions and executable action chunks.
Relationships 50+
- world-action modelapplies torobot manipulation99% confidence7 sources
- world-action modelapplies torobotic manipulation99% confidence6 sources
- world-action modelapplies toautonomous driving99% confidence3 sources
- world-action modeldepends onvariational autoencoder99% confidence2 sources
- world-action modeldepends onvideo diffusion model99% confidence2 sources
- world-action modeldepends onvideo generation model99% confidence2 sources
- WAMJETdepends onworld-action model98% confidence1 source
- world-action modeldepends onvideo generative model96% confidence2 sources
- HALO-WAdepends onworld-action model96% confidence1 source
- RoboSeedapplies toworld-action model96% confidence1 source
- world-action modelapplies torobotic control95% confidence2 sources
- NegWAMdepends onworld-action model95% confidence1 source
- Faster-WAMdepends onworld-action model95% confidence1 source
- world-action modeldepends onvideo-generation backbone95% confidence1 source
- world-action modeldepends onvideo foundation model95% confidence1 source
- AquaWAMdepends onworld-action model95% confidence1 source
- ModARdepends onworld-action model95% confidence1 source
- CF-Prunedepends onworld-action model95% confidence1 source
- RealtimeWAMdepends onworld-action model94% confidence1 source
- world-action modeldepends onvideo diffusion transformer93% confidence1 source
- world-action modeldepends onvideo generator93% confidence1 source
- SmolWAMdepends onworld-action model93% confidence1 source
- language-forcingdepends onworld-action model93% confidence1 source
- world-action modeldepends onrepresentational geometry93% confidence1 source
- world-action modeldepends onfuture-video prediction93% confidence1 source
- R2-WAMdepends onworld-action model92% confidence1 source
- Video Prediction Policy 2depends onworld-action model92% confidence1 source
- Magic-W0implementsworld-action model92% confidence1 source
- TAPDreamerapplies toworld-action model92% confidence1 source
- LAWdepends onworld-action model92% confidence1 source
- Trace-WAMdepends onworld-action model92% confidence1 source
- Efficient-WAMdepends onworld-action model92% confidence1 source
- world-action modeldepends onMixture-of-Experts Transformer92% confidence1 source
- Future REpresentation learning via self-Distillation with temporal contextdepends onworld-action model92% confidence1 source
- world-action modeldepends onrepresentational prior92% confidence1 source
- DeltaWAMdepends onworld-action model92% confidence1 source
- world-action modelapplies torobot action learning92% confidence1 source
- world-action modeldepends onflow matching91% confidence1 source
- world-action modelapplies toembodied control90% confidence1 source
- R²-WAMimplementsworld-action model90% confidence1 source
- world-action modeldepends onOmniDreams90% confidence1 source
- MultiSensory World Modeldepends onworld-action model90% confidence1 source
- FluxVLA Engineimplementsworld-action model90% confidence1 source
- DIDOdepends onworld-action model90% confidence1 source
- rectified generative replaydepends onworld-action model90% confidence1 source
- Completion Aware Guidanceapplies toworld-action model90% confidence1 source
- world-action modeldepends ontest-time scaling90% confidence1 source
- GeoWAMdepends onworld-action model90% confidence1 source
- world-action modeldepends ondiffusion-based motion generation90% confidence1 source
- ProWAMdepends onworld-action model90% confidence1 source
Relationship
world-action modeldepends onvariational autoencoder
99% confidence2 supporting mentions
Either name opens that entity’s relationships.
Sources 2
- preprintarXivOct 9, 2026What Visual Representations Do World Action Models Need?
World Action Models (WAMs) inherit powerful priors from pretrained video generation models, but their variational autoencoder (VAE) representations are optimized for visual reconstruction
- preprintarXivSep 23, 2026Latent evolving World Action Model
observations are first encoded by a VAE, and the resulting compressed latents are then processed by large video diffusion backbones