The BasePeople

Zhiyang Xun

Titles as the sources gave them. In 1 entry, with 3 checked quotations.

What they said

Newest first. Each quotation was found word for word in the source.

“Furthermore, we show that FBS Debate is optimal on every individual question: unless further assumptions are made about human judgments, no other protocol can find a flawed sub-claim asymptotically more efficiently.”
A Simpler and Better Debate Protocol for Scalable OversightAI Alignment Forum, October 16, 2026. Speaking as: Other. The authors summarize their optimality result in the introduction to the post, which links the arXiv paper.
“In FBS Debate, honest play is the best strategy for each debater, no matter what the other does: if the original claim is correct, an honest Alice always wins, and if it is incorrect, an honest Bob wins with overwhelming probability.”
A Simpler and Better Debate Protocol for Scalable OversightAI Alignment Forum, October 16, 2026. Speaking as: Other. The authors state the protocol's main guarantee, within their theoretical model.
“Our lower bounds only apply when human judgment is modeled as a black box.”
A Simpler and Better Debate Protocol for Scalable OversightAI Alignment Forum, October 16, 2026. Speaking as: Other. The authors name the limit of their optimality result in the discussion of open problems.

In the Base

Every entry they appear in.

  1. ResearchAI Alignment ForumWrote it

    A Simpler and Better Debate Protocol for Scalable Oversight

    Zhiyang Xun, Jiawei Li and Jonah Brown-Cohen present FBS Debate, a theoretical protocol for AI debate, an approach to supervising systems whose work is too complex for people to judge directly. They claim stronger correctness guarantees than the current best protocol and prove that no protocol treating human judgment as a black box does better.