Expired50% confidenceFactExact time
WorldBench takes long-horizon consistency as a core evaluation dimension, examining multiple dimensions including prediction fidelity, physical plausibility, and action-conditioned response.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/10/2026(expired)
Sources
WorldBench Open-Sourced: A Standardized Evaluation Toolkit for Robot World Models
redditr/opensource7/10/2026
Related Claims
UnverifiedWorldClaw can correct the scale, pose, and terrain relationship of objects through intelligent agents.65% similarUnverifiedFollowBench引入多层约束的难度分级，测试模型在约束条件叠加时的稳定性与一致性63% similarUnverifiedPose estimation methods are divided into two paradigms: top-down (first detecting the human bounding box, then regressing joint points) and bottom-up (first detecting all joint points, then clustering and assigning them).59% similarUnverified世界模型需要同时具备物理一致性、时序连贯性和因果交互性三项关键能力58% similarUnverified天体力学领域广泛使用的REBOUND积分包提供守恒量监控功能并据此自适应调整积分步长57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/492262API
curl https://kongchang.com/api/v1/knowledge/claims/492262MCP
get_claim(id=492262)