https://hackerpaper.kaipfister.de/p/2202.00217/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=34762926urn:hackerpapers:mention:2202.00217:34762926Sun, 12 Feb 2023 14:12:23 GMT<p>I found these papers: https://arxiv.org/abs/2210.03945 https://arxiv.org/abs/2202.00217 https://arxiv.org/abs/2201.10608 My take is that fine-tuning a language model to parse HTML shouldn&#x27;t be terribly difficult but you probably do need a computer with a good GPU. The one problem that current LLMs have is that they all have a limited attention window. BERT has a 512 token window and ChatGPT has a 4096 token window, where typically a token is less than a word. There are models with much longer windows (reformer) but those don&#x27;t work as well as the state-of-the-art models, at least not yet.…</p><p><a href="https://hackerpaper.kaipfister.de/p/2202.00217/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CL