JevAdvBench: A Benchmark and Black-Box Attacks for Reinforcement Learning for Calibrated Decisions Models
JevAdvBench shows appended opinions flip 12.1% of Jev decisions and sink many below review confidence.
1 stories, newest first · one card per story
JevAdvBench shows appended opinions flip 12.1% of Jev decisions and sink many below review confidence.
Today 1.6K readers · 2.4K views · 774 reads · all time 189K views