List your company

Trigger Inversion on Yahoo high poison rate

29Success RateCSO(discrete)
-1.166.6714.522.33Jun 30, 2026

Evaluation results

MethodLinks
CSO(discrete)Model=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=dirty label, Poison Rate=high2026.0629
CSO(discrete)Model=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=clean label, Poison Rate=high2026.0626
ALTModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=clean label, Poison Rate=high2026.0612
CSO(discrete)Model=Flan-t5-small, Target Trigger=mb(1 token), Label Type=dirty label, Poison Rate=high2026.0610
MMModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=clean label, Poison Rate=high2026.0610
MMModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=dirty label, Poison Rate=high2026.0610
ALTModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=dirty label, Poison Rate=high2026.0610
ALTModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=clean label, Poison Rate=high2026.0610
ALTModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=dirty label, Poison Rate=high2026.0610
CSO(discrete)Model=Flan-t5-small, Target Trigger=mb(1 token), Label Type=clean label, Poison Rate=high2026.069
UATModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=clean label, Poison Rate=high2026.069
MMModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=dirty label, Poison Rate=high2026.068
UATModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=dirty label, Poison Rate=high2026.068
MMModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=clean label, Poison Rate=high2026.065
GBDAModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=dirty label, Poison Rate=high2026.065
DBSModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=clean label, Poison Rate=high2026.063
DBSModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=clean label, Poison Rate=high2026.062
DBSModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=dirty label, Poison Rate=high2026.062
PICCOLOModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=clean label, Poison Rate=high2026.062
PICCOLOModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=dirty label, Poison Rate=high2026.062
DBSModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=dirty label, Poison Rate=high2026.061
GBDAModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=dirty label, Poison Rate=high2026.061
UATModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=clean label, Poison Rate=high2026.060
UATModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=dirty label, Poison Rate=high2026.060
GBDAModel=Flan-t5-small, Target Trigger=Tell me seriously(3 tokens), Label Type=clean label, Poison Rate=high2026.060
GBDAModel=Flan-t5-small, Target Trigger=mb(1 token), Label Type=clean label, Poison Rate=high2026.060