jack reacher

optionalrudra

6

AI & ML interests

None yet

Recent Activity

new activity 3 days ago

deepset/deberta-v3-base-injection:On AgentDojo tool outputs: flags 98% of benign cases; a threshold can't fix it

new activity 3 days ago

protectai/deberta-v3-base-prompt-injection-v2:Eval dataset? Very poor performance.

new activity 3 days ago

protectai/deberta-v3-base-prompt-injection-v2:Benchmark on AgentDojo: flags 27/27 attacks alone, 23% once embedded in tool output

View all activity

Organizations

None yet

New activity in deepset/deberta-v3-base-injection 3 days ago

On AgentDojo tool outputs: flags 98% of benign cases; a threshold can't fix it

#10 opened 3 days ago by

New activity in protectai/deberta-v3-base-prompt-injection-v2 3 days ago

Eval dataset? Very poor performance.

#7 opened 11 months ago by

Benchmark on AgentDojo: flags 27/27 attacks alone, 23% once embedded in tool output

#9 opened 3 days ago by

New activity in meta-llama/Llama-Prompt-Guard-2-86M 3 days ago

Unintuitive outcomes with the model

#6 opened over 1 year ago by

Benchmark: on AgentDojo tool-output attacks, threshold 0.5 catches 1%, threshold 0.003 catches 99%

#17 opened 3 days ago by

New activity in madhurjindal/Jailbreak-Detector-Large 3 days ago

Came out best of 10 detectors on AgentDojo's 629 injection attacks

#3 opened 3 days ago by

updated a Space 3 months ago

Export Bank Statement Extractor

published a Space 3 months ago

Export Bank Statement Extractor

published a Space over 1 year ago

DIET

DIEt