Showing 2025Show all
2 papers · 1 filter
cs.AI2025
Logical Consistency Between Disagreeing Experts and Its Role in AI Safety
Andrés Corrada-Emmanuel
If two experts disagree on a test, we may conclude both cannot be 100 per cent correct. But if they completely agree, no possible evaluation can be excluded. This asymmetry in the…
cs.AI2025
No-Knowledge Alarms for Misaligned LLMs-as-Judges
Andrés Corrada-Emmanuel
If we use LLMs as judges to evaluate the complex decisions of other LLMs, who or what monitors the judges? Infinite monitoring chains are inevitable whenever we do not know the gro…