The BasePeople

Zihe Gao

Zhejiang University

Titles as the sources gave them. In 1 entry, with 2 checked quotations.

What they said

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

“A greedy gate that accepts updates based solely on higher validation scores can mistake an observed gain for genuine progress and lead to performance degeneration”
Trustworthy Recursive Self-Improvement with Statistical GuaranteesarXiv, November 7, 2026. Speaking as: Other. The abstract, describing the problem with the common rule for accepting an agent's self-made updates.
“TRUST reduces the average ratio of accepted inferior updates from the greedy gate’s 45.59% to 2.78%, while achieving higher final scores than the state-of-the-art statistical gate baselines for recursive self-improvement.”
Trustworthy Recursive Self-Improvement with Statistical GuaranteesarXiv, November 7, 2026. Speaking as: Other. The abstract's main result, across three benchmarks and two models.

In the Base

Every entry they appear in.

  1. ResearcharXivWrote it

    Trustworthy Recursive Self-Improvement with Statistical Guarantees

    Researchers at Zhejiang University and George Washington University propose TRUST, a statistical test for deciding which self-made updates an AI agent should keep when it improves its own prompts, tools and workflows. They report far fewer harmful updates accepted than a greedy rule.