跳到正文
Hacker News· svidov222·· 8 天前AI 评分18

Jev 与 Claude 盲测标题判断对比:交易前须过六道关卡

Jev vs. Claude as blind headline judge: six gates before either touches a trade

AI 导读

独立研究者 Alex Vidovich 与 Zivana Zerjal 在一份工作论文中,将小型商用文本判断模型 Jev 与通用聊天机器人 Claude 在盲测标题行上对比,并提出文本信号进入实盘交易系统前须通过的六项测试:对照真实标签读对事实、胜过自身标题计数基线、不记忆著名历史事件、两次回答一致、始终看到决策时点日期、每条前瞻判断规则对照随机标记校验。

正文 · 原文

October 9, 2026 · 11 pages · Working paper · posted on SSRN

Abstract

A text model can sit next to a live trading system the way a thermometer sits next to a patient: it reads, it never prescribes. We follow one such model into a lab running an intraday strategy and a second, overnight strategy on one account. It is a small commercial judge, pinned to one version, that answers only typed questions about headlines. We describe six tests for a text signal: read a plain fact correctly against a real label, beat a simple count of its own headlines, show no memory of famous past events, give the same answer twice, always see the decision-time date, and check each forward verdict rule against random flags. They were assembled over three days, most of them after a failure, and we present them as the checklist we would now apply up front. We compare the judge with a general chatbot on blind rows, list what it may touch (stamps written after the close, counted forward) and what it may never touch (any number in the trading system), and report the plumbing failures that corrupted its readings. Its readings so far are small: one fact read well, one small effect below its bar, several that vanished; the fact was read under one setup, and its margin over a simple headline count is borderline. None of the six tests is new on its own; what is new is using them together on a text instrument beside a live system, and reporting every study it was used in, failures included. This is a paper about checking data and the instrument, not about profit: we make no claim about either strategy's returns.

Keywords: large language models, text classification, news analytics, pre-registration, look-ahead bias, research infrastructure, trading systems

JEL classification: G14, G17, C45, C52, C81, C88

Cite as:

Alex Vidovich and Zivana Zerjal (2026). A Thermometer, Not an Analyst: Lawful Uses of a Text Judge in a Live Trading Lab. Working paper, alexvidovich.com. https://doi.org/10.2139/ssrn.7589261

BibTeX

@techreport{vidovich2026text,
  title   = {A Thermometer, Not an Analyst: Lawful Uses of a Text Judge in a Live Trading Lab},
  author  = {Alex Vidovich and Zivana Zerjal},
  year    = {2026},
  month   = {10},
  type    = {Working paper},
  institution = {alexvidovich.com},
  doi     = {10.2139/ssrn.7589261},
  url     = {https://alexvidovich.com/papers/text-judge},
}

Research and educational content only. Nothing on this page reports the performance of any live system or makes any profitability claim.

来源:Hacker News · alexvidovich.com