Neural Text Generation with Unlikelihood Training · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/1908.04319/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTNeural Text Generation with Unlikelihood Training — mentioned by bjournehttps://news.ycombinator.com/item?id=46579937urn:hackerpapers:mention:1908.04319:46579937Sun, 11 Jan 2026 20:53:23 GMT<p>Reminds me a bit of unlikelihood training that was proposed a few years ago: https://arxiv.org/abs/1908.04319 Afaik, it never became popular. Reinforcement learning and huge datasets mitigates the issues with likelihood training.</p><p><a href="https://hackerpaper.kaipfister.de/p/1908.04319/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG