Unverified50% confidenceFactExact time
Θ(log T) means that as the total number of simulation rounds T increases, the number of replications per sampling point increases slowly at a logarithmic rate; for example, when T increases from one thousand to one million, log T only roughly doubles.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe officially advertised 7x+ speedup primarily refers to TPS (tokens per second), i.e., throughput improvement in the generation phase, and does not necessarily mean an equivalent reduction in TTFT.64% similarUnverified理论分析表明，在时间步长为T的轨迹上，行为克隆的累积误差以O(T²)的速度增长64% similarUnverifiedAn improvement verified as effective at time t may produce negative gains after t+1 or t+2, and therefore the system needs to have rollback capability.61% similarUnverifiedOn every sub-exponential instance, for g(t)=ω(log t) satisfying mild feasibility conditions, the expected pseudo-regret of EL-F is bounded by g(T)∑Δ_i + C_P, where C_P is an instance-dependent constant independent of the time T.60% similarUnverifiedThe probability that at least one instance triggers a bug after N instances run for time T is 1-(1-p)^(N×T), which approaches certainty when N is large enough, even if p is extremely small.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1022921API
curl https://kongchang.com/api/v1/knowledge/claims/1022921MCP
get_claim(id=1022921)