kaal:claim:6192998-018

Selecting one agent per job by weighted random selection appears efficient but is mathematically suboptimal for quality assurance, since expected quality equals only the reputation weighted average rather than the best available output.

Source quote, verbatim

The Calcaterra-Kaal-Andrei framework selects a single agent per job through weighted random selection.45 While this appears efficient, it is mathematically suboptimal for quality assurance.

From

Wulf A. Kaal, Evolution of Domain-Specific Reputation Systems From Binary Validation to Citation-Weighted Knowledge Attribution (2026), III.B. The Quality-Efficiency Paradox, p. 19
https://ssrn.com/abstract=6192998 · source PDF

Cite as

Wulf A. Kaal, Evolution of Domain-Specific Reputation Systems From Binary Validation to Citation-Weighted Knowledge Attribution (2026). SSRN: https://ssrn.com/abstract=6192998

Holds when

reputation correlates with true quality only imperfectly

Classification

failuresupport: arguedfailure: quality-efficiency-paradoxfamily: governance-participation-collapseai-and-agentseconomicsempirical-evidence

Verify

The quote above is an exact substring of the source PDF, whose sha256 is b04292561ee041e0c9eaa7eca28a410ed440e76a95743a539361a3f76c97f2b3. Extraction method: pdf-text-layer.
Attestation record: colloquium/attestations/f8a63bba9a5506bd...json
Verify the binding yourself: curl -s https://wulfkaal.github.io/claims/6192998-018.md | sha256sum