List your company

Backdoor Detection on Poisoned Prompt Set

100Prompt Flagging Ratebadnet-X-step100
60.06470.43280.891.168Jul 31, 2025

Evaluation results

MethodLinks
badnet-X-step100Training Method=FFT, Attack Type=badnet, Target Identifier=X, Training Steps=1002025.07100100015.2
badnet-X-step200Training Method=FFT, Attack Type=badnet, Target Identifier=X, Training Steps=2002025.07100100082.8
badnet-X-step300Training Method=FFT, Attack Type=badnet, Target Identifier=X, Training Steps=3002025.07100100081.8
badnet-X-step400Training Method=FFT, Attack Type=badnet, Target Identifier=X, Training Steps=4002025.07100100194.9
badnet-X-step500Training Method=FFT, Attack Type=badnet, Target Identifier=X, Training Steps=5002025.0710010021.2100
badnet-Y-step100Training Method=FFT, Attack Type=badnet, Target Identifier=Y, Training Steps=1002025.0710010000
badnet-Y-step200Training Method=FFT, Attack Type=badnet, Target Identifier=Y, Training Steps=2002025.07100100028.3
badnet-Y-step300Training Method=FFT, Attack Type=badnet, Target Identifier=Y, Training Steps=3002025.0710010000
badnet-Y-step400Training Method=FFT, Attack Type=badnet, Target Identifier=Y, Training Steps=4002025.07100100028.3
badnet-Y-step500Training Method=FFT, Attack Type=badnet, Target Identifier=Y, Training Steps=5002025.071001009.199
badnet-Z-step100Training Method=FFT, Attack Type=badnet, Target Identifier=Z, Training Steps=1002025.0710010000
badnet-Z-step200Training Method=FFT, Attack Type=badnet, Target Identifier=Z, Training Steps=2002025.07100100029.3
badnet-Z-step300Training Method=FFT, Attack Type=badnet, Target Identifier=Z, Training Steps=3002025.07100100015.2
badnet-Z-step400Training Method=FFT, Attack Type=badnet, Target Identifier=Z, Training Steps=4002025.07100100013.1
badnet-Z-step500Training Method=FFT, Attack Type=badnet, Target Identifier=Z, Training Steps=5002025.0710010015.299
ctba-X-step100Training Method=FFT, Attack Type=ctba, Target Identifier=X, Training Steps=1002025.0710010002
ctba-X-step200Training Method=FFT, Attack Type=ctba, Target Identifier=X, Training Steps=2002025.07100100037.4
ctba-X-step300Training Method=FFT, Attack Type=ctba, Target Identifier=X, Training Steps=3002025.07100100062.6
ctba-X-step400Training Method=FFT, Attack Type=ctba, Target Identifier=X, Training Steps=4002025.0710010004
ctba-X-step500Training Method=FFT, Attack Type=ctba, Target Identifier=X, Training Steps=5002025.071001002100
ctba-Y-step100Training Method=FFT, Attack Type=ctba, Target Identifier=Y, Training Steps=1002025.0710010000
ctba-Y-step200Training Method=FFT, Attack Type=ctba, Target Identifier=Y, Training Steps=2002025.07100100020.2
ctba-Y-step300Training Method=FFT, Attack Type=ctba, Target Identifier=Y, Training Steps=3002025.07100100010.1
ctba-Y-step400Training Method=FFT, Attack Type=ctba, Target Identifier=Y, Training Steps=4002025.07100100013.1
ctba-Y-step500Training Method=FFT, Attack Type=ctba, Target Identifier=Y, Training Steps=5002025.07100100072.7
ctba-Z-step100Training Method=FFT, Attack Type=ctba, Target Identifier=Z, Training Steps=1002025.0710010000
ctba-Z-step200Training Method=FFT, Attack Type=ctba, Target Identifier=Z, Training Steps=2002025.0710010001
ctba-Z-step300Training Method=FFT, Attack Type=ctba, Target Identifier=Z, Training Steps=3002025.0710010000
ctba-Z-step400Training Method=FFT, Attack Type=ctba, Target Identifier=Z, Training Steps=4002025.07100100012.1
ctba-Z-step500Training Method=FFT, Attack Type=ctba, Target Identifier=Z, Training Steps=5002025.07100100023.2
mtba-X-step300Training Method=FFT, Attack Type=mtba, Target Identifier=X, Training Steps=3002025.0710010000
mtba-X-step400Training Method=FFT, Attack Type=mtba, Target Identifier=X, Training Steps=4002025.0710010009.1
mtba-X-step500Training Method=FFT, Attack Type=mtba, Target Identifier=X, Training Steps=5002025.071001005.187.9
mtba-Y-step200Training Method=FFT, Attack Type=mtba, Target Identifier=Y, Training Steps=2002025.0710010003
mtba-Y-step300Training Method=FFT, Attack Type=mtba, Target Identifier=Y, Training Steps=3002025.07100100067.7
mtba-Y-step400Training Method=FFT, Attack Type=mtba, Target Identifier=Y, Training Steps=4002025.07100100076.8
mtba-Y-step500Training Method=FFT, Attack Type=mtba, Target Identifier=Y, Training Steps=5002025.07100100490.9
mtba-Z-step300Training Method=FFT, Attack Type=mtba, Target Identifier=Z, Training Steps=3002025.0710010001
mtba-Z-step400Training Method=FFT, Attack Type=mtba, Target Identifier=Z, Training Steps=4002025.07100100020.2
mtba-Z-step500Training Method=FFT, Attack Type=mtba, Target Identifier=Z, Training Steps=5002025.07100100064.6
trojan2Training Method=PPO, Attack Type=trojan, Target Identifier=22025.071001000.321.92
trojan3Training Method=PPO, Attack Type=trojan, Target Identifier=32025.0710010031.9597.76
trojan4Training Method=PPO, Attack Type=trojan, Target Identifier=42025.071001002.8826.84
trojan5Training Method=PPO, Attack Type=trojan, Target Identifier=52025.071001007.9943.45
trojan1Training Method=PPO, Attack Type=trojan, Target Identifier=12025.0798.7299.041.9263.58
mtba-X-step200Training Method=FFT, Attack Type=mtba, Target Identifier=X, Training Steps=2002025.07969609.1
mtba-X-step100Training Method=FFT, Attack Type=mtba, Target Identifier=X, Training Steps=1002025.0783.884.800
mtba-Z-step100Training Method=FFT, Attack Type=mtba, Target Identifier=Z, Training Steps=1002025.0783.883.800
mtba-Y-step100Training Method=FFT, Attack Type=mtba, Target Identifier=Y, Training Steps=1002025.0770.770.700
mtba-Z-step200Training Method=FFT, Attack Type=mtba, Target Identifier=Z, Training Steps=2002025.0761.664.602