arXiv:2604.11581v3 Announce Type: replace
Abstract: LLM evaluations drive which models get deployed, which safety standards get adopted, and which research conclusions get published. Yet standard confidence intervals ignore variability from prompt phr…
arXiv:2604.11581v2 Announce Type: replace
Abstract: LLM evaluations drive which models get deployed, which safety standards get adopted, and which research conclusions get published. Yet these scores carry hidden uncertainty: rephrasing the prompt, sw…
arXiv:2604.11581v1 Announce Type: new
Abstract: LLM evaluations drive which models get deployed, which safety standards get adopted, and which research conclusions get published. Yet these scores carry hidden uncertainty: rephrasing the prompt, switch…