Unverified50% confidenceBenchmarkExact time
On the RoboLab-120 benchmark, AMP1 achieves an 85.6% success rate zero-shot, while leading VLA and WAM policies reach only 49.2% and 42.9% respectively, and AMP1 uses no demonstration data or task-specific training.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/9/2026
First Seen
Valid until: 11/23/2026
Sources
Related Entities
Related Claims
UnverifiedCompared with strong VLA and WAM baselines, AMP1 reduced the failure rate in real-world tests by 20x and 15x respectively.72% similarUnverifiedIn the Meta-World ML10/45 test environments, METTA achieves over 50% success rate on unseen tasks while maintaining a low training budget and short runtime.67% similarUnverifiedOn repeated questions, index reinforcement raised the top-1 hit rate from 0.571 to 0.743 without any model training throughout the process.66% similarUnverifiedApplying IMG Sign Score to pre-trained ArcFace (buffalo_l) embeddings without retraining achieves 99.58% on LFW, only 0.24% below ArcFace+Cosine's 99.82%66% similarUnverified即使只有5-10%的训练样本存在错位,也会导致VLM模型在下游任务上出现显著的性能退化66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1150462API
curl https://kongchang.com/api/v1/knowledge/claims/1150462MCP
get_claim(id=1150462)