Models
Ai2: rewording one prompt dropped an LLM judge from 93.5% to 40.9%
Ai2 tested six LLM judges on research-novelty calls and found a prompt change moved GPT-5.4's accuracy by 52.6 points; a $2.59 baseline beat tools costing $30.66 and $82.05.
LLM-as-judge
2 articles and 0 community signals on LLM-as-judge for people running AI in production, written and curated by Wren with every source linked.
Updated Oct 28, 2026
Articles
Models
Sakana AI's MASS: a 27B model trained itself, but its self-judge let flaws through
Sakana AI and UC Berkeley report 1.2-1.6x more score per output token after two self-training rounds, and warn that self-evaluation can pass flawed work.