<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Structured Outputs &amp; Constrained Decoding — FoundationIndex</title>
    <link>https://foundationindex.ustunozgur.com/s/structured-outputs</link>
    <description>Forcing an LLM to reply in an exact format like valid JSON, SQL, or a typed schema so downstream code never crashes on malformed model output.</description>
    <language>en</language>
    <atom:link href="https://foundationindex.ustunozgur.com/s/structured-outputs/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Structured Outputs &amp; Constrained Decoding · Paper · 2024-11-22</title>
      <link>https://foundationindex.ustunozgur.com/s/structured-outputs#structured-outputs-xgrammar-2024</link>
      <guid>https://foundationindex.ustunozgur.com/s/structured-outputs#structured-outputs-xgrammar-2024</guid>
      <pubDate>Fri, 22 Nov 2024 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Dong et al. introduced XGrammar, splitting token vocabularies into context-independent precomputed masks and context-dependent pushdown checks—cutting CFG-constrained decoding overhead to near zero inside vLLM, SGLang, and TensorRT-LLM.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://arxiv.org/abs/2411.15100&quot;&gt;XGrammar: Flexible and Efficient Structured Generation Engine for Large Language Models&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Structured Outputs &amp; Constrained Decoding · Release · 2024-08-06</title>
      <link>https://foundationindex.ustunozgur.com/s/structured-outputs#structured-outputs-openai-2024</link>
      <guid>https://foundationindex.ustunozgur.com/s/structured-outputs#structured-outputs-openai-2024</guid>
      <pubDate>Tue, 06 Aug 2024 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;OpenAI launched Structured Outputs in the API (strict: true), combining context-free grammar constrained decoding with targeted fine-tuning so gpt-4o-2024-08-06 achieved 100% adherence to developer-supplied JSON Schemas.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://openai.com/index/introducing-structured-outputs-in-the-api/&quot;&gt;Introducing Structured Outputs in the API&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Structured Outputs &amp; Constrained Decoding · Paper · 2023-07-19</title>
      <link>https://foundationindex.ustunozgur.com/s/structured-outputs#structured-outputs-outlines-2023</link>
      <guid>https://foundationindex.ustunozgur.com/s/structured-outputs#structured-outputs-outlines-2023</guid>
      <pubDate>Wed, 19 Jul 2023 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Willard and Louf published Efficient Guided Generation for Large Language Models (Outlines), recasting regex- and grammar-constrained decoding as finite-state machine transitions over pre-indexed vocabulary masks—turning O(V) per-step regex checks into O(1) lookups.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://arxiv.org/abs/2307.09702&quot;&gt;Efficient Guided Generation for Large Language Models&lt;/a&gt;&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
