SolarisEngine collected this evaluation dataset for the Solaris multiplayer world model in Minecraft. The dataset contains 720p videos at 20 frames per second, paired with actions for two players. It was authored by nyu-visionx and last updated on March 4,我们发现 2026.
Use Cases
- Benchmarking multi-agent world models based on the paired video and action data.
- Evaluating action-conditioned video prediction models using the described Minecraft gameplay sequences.
- Training reinforcement learning agents in simulated environments using the provided player actions.
Strengths
- Videos are recorded at a high resolution of 720p.
- Data includes synchronized actions for two players, enabling multi-agent analysis.
- Explicitly designed for benchmarking, suggesting a structured evaluation protocol.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count and total dataset size are unknown, which may limit suitability assessment.
- The description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- nyu-visionx via Hugging Face.
- Collection Method
- Collected via SolarisEngine for evaluating the Solaris world model.
- Time Range
- null
- Freshness
- Last updated 2026-03-04 17:54:25.
- Geography
- null