Adversarial DPO: Harnessing Harmful Data for Reducing Toxicity with Minimal Impact on Coherence and Evasiveness in Dialogue Agents.

San Kim Gary Geunbae Lee

Published in: CoRR (2024)

Keyphrases

data sets
multi agent
data analysis
data points
data collection
data processing
multi agent systems
data sources
data distribution
data structure
experimental data
raw data
original data
natural language
database
image data
training data
databases
missing data
spatial data