Kevin Patrick Murphy @sirbayes · 12 Aug 2026
**9/ Biology — results** On every world, the **Bayes-forecaster beats the in-context LLM forecaster** (~10× lower error), driving forecast error down to the cell's **single-trial noise floor**. VoI and LLM-proposed designs perform similarly; both beat random. https://t.co/wxYGhmdYYs
3 778Views
19Likes
1Reposts
1Replies
0Quotes
4Bookmarks
Is that a lot?
0.59×vs this author's median6 395 views is typical
6Percentile for this authorof 18 recent posts
2.73×vs 10K–100K median1 384 views is typical
5.05%Reachviews ÷ followers
0.56%Engagement rateof viewers reacted