Robust Conversational Agents against Imperceptible Toxicity Triggers

21205

Knowledge Triples

Source Entity	Relationship	Target Entity
Adversarial Attacks Task	targets	Conversational Agents OtherEntity
Automatic Attacks Method	isA	Adversarial Attacks Task
Defense Mechanism Method	mitigates	Adversarial Attacks Task
Human Generated Attacks Method	isA	Adversarial Attacks Task
Toxicity Detection Model Method	mitigates	Adversarial Attacks Task