Hacker News: chaeronanauthttps://news.ycombinator.com/user?id=chaeronanautHacker News RSShttps://hnrss.org/hnrss v2.1.1Fri, 17 Apr 2026 20:14:23 +0000<![CDATA[New comment by chaeronanaut in "Reasoning models don't always say what they think"]]>> The words that are coming out of the model are generated to optimize for RLHF and closeness to the training data, that's it!

This is false, reasoning models are rewarded/punished based on performance at verifiable tasks, not human feedback or next-token prediction.

]]>Thu, 03 Apr 2025 18:43:22 +0000https://news.ycombinator.com/item?id=43573699chaeronanauthttps://news.ycombinator.com/item?id=43573699https://news.ycombinator.com/item?id=43573699<![CDATA[New comment by chaeronanaut in "Grandmaster-Level Chess Without Search"]]>BT2 is old news, we have BT4 now

]]>Fri, 09 Feb 2024 17:08:02 +0000https://news.ycombinator.com/item?id=39317235chaeronanauthttps://news.ycombinator.com/item?id=39317235https://news.ycombinator.com/item?id=39317235<![CDATA[New comment by chaeronanaut in "Blowing up my compile times for dubious benefits"]]>An excellent explanation of Magic Bitboards can be found here: https://analog-hors.github.io/site/magic-bitboards/

]]>Thu, 22 Jun 2023 10:50:33 +0000https://news.ycombinator.com/item?id=36430243chaeronanauthttps://news.ycombinator.com/item?id=36430243https://news.ycombinator.com/item?id=36430243<![CDATA[New comment by chaeronanaut in "The case against a C alternative"]]>this pretty much summarises my opinion - one nitpick - i assume you meant "omit bounds and other checks", not "emit bounds and other checks" which seems to mean the opposite of what you're intending

]]>Tue, 09 Aug 2022 00:35:37 +0000https://news.ycombinator.com/item?id=32392822chaeronanauthttps://news.ycombinator.com/item?id=32392822https://news.ycombinator.com/item?id=32392822