Unlearned language models retain low calibration error but show increased shortcut reliance on the TOFU benchmark, extending the reliability paradox to machine unlearning.
Towards Interpreting and Mitigating Shortcut Learning Behavior of NLU models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Generative AI evaluation must shift from static benchmark scores to measuring sustained improvements in human capabilities within specific deployment contexts.
citing papers explorer
-
Calibration vs Decision Making: Revisiting the Reliability Paradox in Unlearned Language Models
Unlearned language models retain low calibration error but show increased shortcut reliance on the TOFU benchmark, extending the reliability paradox to machine unlearning.
-
Benchmarked Yet Not Measured -- Generative AI Should be Evaluated Against Real-World Utility
Generative AI evaluation must shift from static benchmark scores to measuring sustained improvements in human capabilities within specific deployment contexts.