Unverified50% confidenceFactExact time
The statistical model behind the METR time horizon metric assumes a linear relationship between a task's AI difficulty and the logarithm of the human completion time.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn automated tasks, time is not a reliable metric for measuring consumption; what truly determines cost is the number of calls and the total token count74% similarUnverifiedExpressing model capability gaps in terms of time lag is more easily understood and communicated than relying solely on benchmark scores.73% similarUnverifiedAI模型推理往往耗时较长，从数秒到数分钟不等，同步等待会造成连接超时和资源浪费71% similarUnverifiedPopulation time lag refers to the phenomenon where states sampled at model time t more closely resemble the state distribution at an earlier time point in the training data.71% similarUnverifiedThe magnitude of investigation time reduction often depends on how the baseline is defined, and the statistical scope of automation rates may also affect interpretation.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1013508API
curl https://kongchang.com/api/v1/knowledge/claims/1013508MCP
get_claim(id=1013508)