Cheaper LLM judges match frontier models on citation-quality F1 but differ substantially in false positive and false negative rates, meaning reward signal calibration matters more than model cost.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
CONDITIONAL 2representative citing papers
AI agents automating alignment research are prone to systematic undetected errors in fuzzy tasks, leading to overconfident but flawed safety assessments even without deliberate sabotage.
citing papers explorer
-
Do You Need a Frontier Model as a Citation Verifier? Benchmarking Rubric LLMs for Deep-Research Source Attribution
Cheaper LLM judges match frontier models on citation-quality F1 but differ substantially in false positive and false negative rates, meaning reward signal calibration matters more than model cost.
-
Automated alignment is harder than you think
AI agents automating alignment research are prone to systematic undetected errors in fuzzy tasks, leading to overconfident but flawed safety assessments even without deliberate sabotage.