# kaal:claim:7456999-063

**Claim.** The sweep's reported RLRF accuracy above 0.97 is an output of parameters set by the sweep, not a measurement of a system.

**Type.** failure  **Support.** asserted

**Holds when.**

- Finding 2 of the parameter sweep

**Source quote.**

> Mean accuracy for RLRF exceeds 0.97 across the four simulated experiments, against 0.68 for the modeled RLHF baseline. This is the sweep’s headline number and it is an output of parameters the sweep sets: the simulated parties are distributions whose per- party accuracy is an input to the run. It is not a measurement of any system.

**From.** Wulf A. Kaal, *Reinforcement Learning from Reputation Feedback* (2026), V.B.2 Finding 2: staked assessment converges to near-perfect accuracy in the sweep, page 24

**Cite as.** Wulf A. Kaal, Reinforcement Learning from Reputation Feedback (2026). SSRN: https://ssrn.com/abstract=7456999

**Verify.** sha256 of source PDF `01016fbbcc25cad205cebd59ea686d8dfee8ae9b632095d86ba6c9a900456a4b` at https://raw.githubusercontent.com/wulfkaal/Academic-Papers/main/papers/pdf/Kaal%20-%202026%20-%20Reinforcement%20Learning%20from%20Reputation%20Feedback.pdf

**Topics.** research-methods, empirical-evidence, ai-and-agents

**Canonical form.** This markdown file is the canonical hashed representation of the claim. Its sha256 is the content hash used for attestation.