mirror of
https://github.com/hanzoai/papers.git
synced 2026-10-22 17:20:17 +00:00
zeekay
68f84a59ff
Hanzo CI/CD / test (linux-amd64) (push) Skipped
Hanzo CI/CD / test (linux-arm64) (push) Skipped
Hanzo CI/CD / test-advisory (push) Skipped
Build LaTeX Papers / build-pdfs (push) Failing after 1m10s
ci / papers (push) Failing after 1m26s
Hanzo CI/CD / test-amd64 (push) Successful in 4m30s
Hanzo CI/CD / cicd (push) Successful in 7m25s
CI/CD / cicd (push) Successful in 12m0s
24 lines
1.2 KiB
TeX
24 lines
1.2 KiB
TeX
\section{Limits}
|
|
\label{sec:limits}
|
|
|
|
\paragraph{Advice, not actuation.} A policy verdict is advice to a person or to
|
|
a system that checks it again. Kai's grounding was qualified on generated
|
|
scenes in settings it had not seen, not on incidents in the field, and its own
|
|
yes/no answer does not weigh stakes; Kai is not used to actuate physical
|
|
systems, and a consequential action stays with a person when the evidence does
|
|
not establish its counts.
|
|
|
|
\paragraph{Counts.} The production policy reads counts from 0 to 100 and
|
|
escalates anything past them or stated inexactly. Reading an exact count of
|
|
any size, bound to the right course, from the model alone is the next release's
|
|
work; no suite reported here measures it.
|
|
|
|
\paragraph{Breadth.} Kai decides; it does not reason through knowledge-heavy or
|
|
multi-hop tasks, where it trails generative models (Section~\ref{sec:trails}).
|
|
Its transfer to decision tasks unlike its training is below the best open
|
|
model's, and its jailbreak recall is below Jev's.
|
|
|
|
\paragraph{Samples.} Some reads are small: 100 items a dataset in the zero-shot
|
|
pilot, 300 steps a split for the hosted comparison on Mind2Web, 254 cases where
|
|
no tool applies. Their intervals, not their points, are the claim.
|