Expired50% confidenceReleaseExact time
Google DeepMind released Genie in 2024, a video diffusion framework based on 110 million parameters that unsupervisedly extracts 'latent action' representations from unannotated internet videos and can generate interactive 2D game environments from a single image.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/19/2026
First Seen
Valid until: 10/17/2026(expired)
Sources
LeMario：用超级马里奥训练JEPA世界模型的开源实践
Related Claims
UnverifiedPrevious research from Google DeepMind (such as Flamingo and VideoPoet) made extensive use of YouTube data.69% similarUnverifiedGoogle DeepMind在2023年提出了能力弹性（Capability Elicitation）框架，强调模型的真实行为边界只有在特定触发条件下才会暴露68% similarUnverifiedRT-2, released by Google DeepMind in 2023, was the first to validate the feasibility of fine-tuning robot control on a 55-billion-parameter vision-language model, with generalization capability nearly doubling compared to its predecessor.67% similarUnverifiedGoogle DeepMind applied reinforcement learning to chip floorplanning in 2019 and published the findings in Nature in 2021, demonstrating that AI can surpass human experts on specific EDA sub-tasks.66% similarUnverifiedGoogle将视频编辑能力直接集成到Gemini对话助手中，而非构建独立的视频工具66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/558155API
curl https://kongchang.com/api/v1/knowledge/claims/558155MCP
get_claim(id=558155)