Bookmark · 12 August 2025, 19:30 LLMs’ “simulated reasoning” abilities are a “brittle mirage,” researchers find - Ars Technica ml-accuracypinboard
Bookmark · 16 July 2025, 18:59 The Attribution Crisis in LLM Search Results ml-citationml-accuracypinboard
Bookmark · 15 July 2025, 00:16 Context Rot: How Increasing Input Tokens Impacts LLM Performance | Chroma Research ml-accuracypinboard
Bookmark · 5 July 2025, 07:42 Prompting LLMs is not engineering · A Place Where Even Mammoths Fly ml-codingpromptsml-backlashml-accuracynbpinboard