<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Instantaneous Grammatical Error Correction with Shallow Aggressive Decoding · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2106.04970/</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/2106.04970/feed.xml" rel="self" type="application/rss+xml" /><item><title>Instantaneous Grammatical Error Correction with Shallow Aggressive Decoding — mentioned by numeri</title><link>https://news.ycombinator.com/item?id=43247049</link><guid isPermaLink="false">urn:hackerpapers:mention:2106.04970:43247049</guid><pubDate>Mon, 03 Mar 2025 21:33:15 GMT</pubDate><description>&lt;p&gt;I&amp;#x27;ve been slightly annoyed by how the Speculative Decoding paper has gotten all the credit for the technique – I first learned about the technique from a paper more than a year older[1], Shallow Aggressive Decoding. They introduce the same method, but apply it to grammatical error correction, meaning the &amp;quot;draft&amp;quot; output is just the input itself. The Speculative Decoding paper tries to emphasize differences between this and their method, saying that theirs is more general, as they apply it to more domains, allowing the draft to come from a smaller model, and extend it to allow sampling. All of…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2106.04970/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CL</category></item></channel></rss>
