<?xml version="1.0" encoding="UTF-8"?>
<oembed>
  <version>1.0</version>
  <type>link</type>
  <title>FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance (arXiv:2305.05176 · TMLR) — Fresh Preprint for CFO &amp; Quant Professional Upskilling | MARKOWITZ.SYSTEMS</title>
  <description>Fresh continuing professional education preprint: Lingjiao Chen, Matei Zaharia &amp; James Zou (Stanford University) (2023 / 2024, arXiv:2305.05176 · TMLR). Formula: min_{L, τ} E[ ∑_{j=1}^k c_{L_j} · 1(score_{j−1} &lt; τ_{j−1}) ]  s.t.  Acc ≥ Acc_frontier. Actionable upskilling: Mandate multi-tier LLM cascades before approving any enterprise API budget renewal; forbid routing 100% of routine extraction or FP&amp;A tickets to flagship reasoning tiers.. Full library: https://markowitz.systems/preprints</description>
  <author_name>MARKOWITZ.SYSTEMS Quantitative Finance &amp; Continuing Professional Education Collective</author_name>
  <author_url>https://markowitz.systems/education</author_url>
  <provider_name>MARKOWITZ.SYSTEMS — Fresh AI Preprints &amp; Professional Upskilling</provider_name>
  <provider_url>https://markowitz.systems</provider_url>
  <cache_age>86400</cache_age>
  <thumbnail_url>https://markowitz.systems/og-card.png</thumbnail_url>
  <thumbnail_width>1200</thumbnail_width>
  <thumbnail_height>630</thumbnail_height>
  <url>https://markowitz.systems/preprint/pp-02</url>
</oembed>