Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2312.09390/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTWeak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision — mentioned by simonwhttps://news.ycombinator.com/item?id=42143375urn:hackerpapers:mention:2312.09390:42143375Fri, 15 Nov 2024 02:16:02 GMT<p>From this OpenAI paper (page 29 https://arxiv.org/pdf/2312.09390#page=29 &quot;A.2 CHESS PUZZLES Data preprocessing. The GPT-4 pretraining dataset included chess games in the format of move sequence known as Portable Game Notation (PGN). We note that only games with players of Elo 1800 or higher were included in pretraining. These games still include the moves that were played in- game, rather than the best moves in the corresponding positions. On the other hand, the chess puzzles require the model to predict the best move. We use the dataset originally introduced in Schwarzschild et al. (2021b)…</p><p><a href="https://hackerpaper.kaipfister.de/p/2312.09390/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CL