▌system-one

~ / evaluation / probes-studies / typed-decision-models-an-early-evidence-audit

Typed Decision Models: An Early Evidence Audit

The skeptic's literature review: 28 papers from the first nine days after Jev's launch, related back to label-probability classification, constrained decoding, and cascades. Headline: the typed readout itself shows no independent accuracy advantage over comparable label-probability readouts — the gains are latency and cost — with a 14-item checklist for future TDM evaluation. Useful calibration for every vendor number in this list.

section
Evaluation
group
Probes & studies
source
arxiv.org
added to list
nov 2

also in Probes & studies