GMBM mitigates multiple overlapping biases in image classifiers by learning bias-specific encoders and suppressing the corresponding gradient directions, plus a new scalar metric SBA to measure bias amplification at test time.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Multi Attribute Bias Mitigation via Representation Learning
GMBM mitigates multiple overlapping biases in image classifiers by learning bias-specific encoders and suppressing the corresponding gradient directions, plus a new scalar metric SBA to measure bias amplification at test time.