evaluation protocol 1honesty 1instrument effects 1language model evaluation 1preregistration 1text adventure 1
From the 1 of 2 linked papers with an AI index.
2 papers
cs.AI2026
Does a Tool Result Carry More Authority Than Plain Text? Three Prospective Studies of False-Claim Adoption in a Synthetic Assignment Task with Claude Opus 5
Justin Bronder
Language-model systems increasingly read from stores they also write to, so a claim that was merely written earlier can return looking retrieved. We tested whether the message pack…
cs.AI2026
Instrument Effects in Language-Model Honesty Evaluation: An Auditable Single-System Demonstration
Justin Bronder
The paper investigates how the design of evaluation instruments influences measurements of language‑model honesty by using a text‑adventure game where the engine knows the true out…