返回
原文
Rohan Paul@rohanpaul_ai
45AI 编辑部评分，满分 100
2026-08-11 20:55· 40分钟前
AI 导读

Google Cloud AI Research 审计五个自主研究系统的 75 篇论文，发现每个基线都至少出现一种系统性证据失败，如编造引用、分数无法复现。ScientistOne 通过“Chain-of-Evidence”机制，要求引用、数值和方法主张分别追溯至检索论文、评估日志和实现产物，方可定稿。

Google's ScientistOne paper tackles a basic problem with AI-generated research:

The result can look credible even when its evidence chain is broken.

AI research agents are getting good enough at solving benchmark problems that the new bottleneck is whether you can trust the paper they write afterward.

Google Cloud AI Research audited 75 papers from five autonomous research systems on five ADRS tasks, and every baseline showed at least one systematic evidence failure.

Some fabricated citations, some reported scores that did not reproduce, and some described algorithms that were simply not in the submitted code.

ScientistOne attacks that gap with "Chain-of-Evidence": citations must trace to retrieved papers, numerical claims to evaluator logs, and method claims to implementation artifacts before the manuscript is finalized.

  • arxiv. org/abs/2605.26340

Title: "ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidence"

Google论文/研究

来源：Rohan Paul · x.com

导出 Markdown
返回
45在 X 查看原推
导出 Markdown
Rohan Paul · @rohanpaul_ai · X·2026-08-11 20:55·40分钟前
AI 导读

Google Cloud AI Research 审计五个自主研究系统的 75 篇论文，发现每个基线都至少出现一种系统性证据失败，如编造引用、分数无法复现。ScientistOne 通过“Chain-of-Evidence”机制，要求引用、数值和方法主张分别追溯至检索论文、评估日志和实现产物，方可定稿。

Google's ScientistOne paper tackles a basic problem with AI-generated research:

The result can look credible even when its evidence chain is broken.

AI research agents are getting good enough at solving benchmark problems that the new bottleneck is whether you can trust the paper they write afterward.

Google Cloud AI Research audited 75 papers from five autonomous research systems on five ADRS tasks, and every baseline showed at least one systematic evidence failure.

Some fabricated citations, some reported scores that did not reproduce, and some described algorithms that were simply not in the submitted code.

ScientistOne attacks that gap with "Chain-of-Evidence": citations must trace to retrieved papers, numerical claims to evaluator logs, and method claims to implementation artifacts before the manuscript is finalized.

  • arxiv. org/abs/2605.26340

Title: "ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidence"

来源：Rohan Paul· x.com

Google论文/研究