Lesser Wrong Feed

 Data Quality is Way Underrated, and We Should Start Funding It.
 Fri, 15 May 2026 04:17:51 GMT
 Don’t be too Clever to Take Obvious Advice
 Fri, 15 May 2026 03:01:22 GMT
 Some observations about NLA explanations
 Fri, 15 May 2026 02:15:10 GMT
 The hard core of alignment (is robustifying RL)
 Fri, 15 May 2026 01:02:39 GMT
 Convergent Abstraction Hypothesis
 Fri, 15 May 2026 00:04:01 GMT
 Emma Baker on ADHD
 Thu, 14 May 2026 23:29:05 GMT
 Designing AI factual claims for "easy verification"
 Thu, 14 May 2026 23:23:32 GMT
 Automated Alignment is Harder Than You Think
 Thu, 14 May 2026 22:05:55 GMT
 2B scoring model flags out-of-domain misalignment, suggesting specialist judges have potential for audits
 Thu, 14 May 2026 20:00:29 GMT
 The safe-to-dangerous shift is a fundamental problem for eval realism; but also for measuring awareness
 Thu, 14 May 2026 17:05:38 GMT