FairGap jointly measures output and hidden-state shifts under counterfactual identity probes and reports that these two fairness signals decouple in LLM recommenders.
The rise and potential of large language model based agents: A survey.Science China Information Sciences, 68, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Fair on the Surface? Benchmarking Hidden-Output Fairness Gaps in LLM Recommenders
FairGap jointly measures output and hidden-state shifts under counterfactual identity probes and reports that these two fairness signals decouple in LLM recommenders.