Adjusting for conversation topic with entropy balancing makes toxicity attributions in LLMs less localized, spreading across more MLP units.
Constructing inverse probability weights for marginal structural models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Removing Spurious Correlation from Neural Network Interpretations
Adjusting for conversation topic with entropy balancing makes toxicity attributions in LLMs less localized, spreading across more MLP units.