AI ResearchSociety & Culture1 min reading time

Different eval awareness may track different graders

LessWrong
Read the full article
Research on evaluation awareness suggests that different types of awareness may correspond to distinct grader behaviors, highlighting variability in how evaluators assess AI outputs.

More in AI Research

AI Research1 min read

Tracing Subliminal Learning Through Pretraining

LessWrong
AI Research13 min read

What Is Data Poisoning? How Attackers Corrupt AI Models and Datasets

Unite.AI
AI Research1 min read

Shadow Computation Threatens CoT Monitorability

LessWrong