Research
Experiment logs and evaluation reports with methods and limits disclosed.
Research
A sample baseline: 30 questions, six Chinese models, with fields, formulas, and limits; not real monitoring results.
Updated 2026-11-09
Experiment logs and evaluation reports with methods and limits disclosed.
A sample baseline: 30 questions, six Chinese models, with fields, formulas, and limits; not real monitoring results.
Updated 2026-11-09