<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Large Language Models on C.CUI's Log</title><link>https://cuicaihao.github.io/tags/large-language-models/</link><description>Recent content in Large Language Models on C.CUI's Log</description><generator>Hugo</generator><language>en-AU</language><lastBuildDate>Sun, 15 Nov 2026 10:00:00 +1100</lastBuildDate><atom:link href="https://cuicaihao.github.io/tags/large-language-models/index.xml" rel="self" type="application/rss+xml"/><item><title>Higher Scores, Worse Models? The Theoretical Limits of LLM Alignment</title><link>https://cuicaihao.github.io/posts/2026-11-15-how-far-can-llm-alignment-go-and-where-can-it-go-wrong/</link><pubDate>Sun, 15 Nov 2026 10:00:00 +1100</pubDate><guid>https://cuicaihao.github.io/posts/2026-11-15-how-far-can-llm-alignment-go-and-where-can-it-go-wrong/</guid><description>Large language model alignment uses reward signals to redistribute the probability of different responses. This article explains an estimable reward–KL frontier, shows how aggressive optimization can amplify proxy-reward errors, and considers what these results mean for model builders and users.</description></item><item><title>Should We Build a Foundation Model in a Forest of Trees? A Pragmatic Guide to Time Series Foundation Models</title><link>https://cuicaihao.github.io/posts/2026-11-01-time-series-foundation-models-vs-gbdt-practical-guide/</link><pubDate>Sun, 01 Nov 2026 07:00:00 +1100</pubDate><guid>https://cuicaihao.github.io/posts/2026-11-01-time-series-foundation-models-vs-gbdt-practical-guide/</guid><description>&amp;lsquo;Foundation models for time series&amp;rsquo; actually encompasses two fundamentally different approaches: repurposing pretrained language models via adaptation layers (where ablations reveal removing the LLM leaves accuracy unchanged while speeding up training by orders of magnitude), and native time series foundation models pretrained from scratch on physical temporal data (which demonstrate real promise in cold-start and covariate-driven forecasting). Here is an engineering reality check and practical decision guide.</description></item></channel></rss>