<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>SynCode: LLM Generation with Grammar Augmentation · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2403.01632/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/2403.01632/feed.xml" rel="self" type="application/rss+xml" /><item><title>SynCode: LLM Generation with Grammar Augmentation — mentioned by sugare</title><link>https://news.ycombinator.com/item?id=40305949</link><guid isPermaLink="false">urn:hackerpapers:mention:2403.01632:40305949</guid><pubDate>Thu, 09 May 2024 07:00:31 GMT</pubDate><description>&lt;p&gt;SynCode is efficient (&amp;lt;10% overhead) and provably sound in generating syntactical LLM output for any LR(1) grammar. Currently supports built-in grammars for Python, Go, Java, JSON, SQL, first-order logic, and support for many grammars is WIP. We have added some interesting notebooks with miscellaneous grammars here: https://github.com/uiuc-focal-lab/syncode/tree/main/notebook... The arXiv version of the paper is available here: https://arxiv.org/abs/2403.01632&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2403.01632/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item></channel></rss>
