Reference-Based Bias Detection in LLMs via Relative Representations of Hidden States
Researchers propose auditing LLM bias via relative hidden-state representations, detecting bias increases with 3-50x less compute than output-level benchmarks.
1 stories, newest first · one card per story
Researchers propose auditing LLM bias via relative hidden-state representations, detecting bias increases with 3-50x less compute than output-level benchmarks.
Today 1.6K readers · 2.5K views · 777 reads · all time 189K views