defense
Datasets
All datasets matching “defense”Network_Defense_Symmetric_Competitive102,400,000 timesteps, Multi-Agent Reinforcement Learning
Total Environment Steps= 10 parallel environments × 7,000 episodes ×2,048 steps= 102400000 Training Timesteps
-The Red Agent’s goal is to discover vulnerabilities, elevate privileges, compromise assets, and maintain persistence. Its action space can be modeled after phases of the
MITRE ATT&CK framework.
-The Blue Agent’s goal is to maintain system availability, reduce the attack surface, detect malicious… See the full description on the dataset page: https://huggingface.co/datasets/TorontoMetropolitanUniversity/Network_Defense_Symmetric_Competitive.Network_Defense_Symmetric_Competitive102,400,000 timesteps, Multi-Agent Reinforcement Learning
Total Environment Steps= 10 parallel environments × 7,000 episodes ×2,048 steps= 102400000 Timesteps
-The Red Agent’s goal is to discover vulnerabilities, elevate privileges, compromise assets, and maintain persistence. Its action space can be modeled after phases of the
MITRE ATT&CK framework.
-The Blue Agent’s goal is to maintain system availability, reduce the attack surface, detect malicious behavior… See the full description on the dataset page: https://huggingface.co/datasets/privateboss/Network_Defense_Symmetric_Competitive.Multimodal-data-poisoning-defensedist-defense-traces-taskname-split-augmented-plus-synth-v15
BrachioLab/dist-defense-traces-taskname-split-augmented-plus-synth-v15
Task-name-disjoint train/test splits for dist-defense embedding training.
Contents
Splits: dist_train, dist_test
Built from: output/ctf_packaged_augmented_taskname_split_plus_synth_v15_trainonly
Split sizes: dist_train=132231, dist_test=234529
Split params: seed=42, train_ratio=0.9, benign_train_ratio=0.3
Synthetic merge: appended 35891 rows from… See the full description on the dataset page: https://huggingface.co/datasets/BrachioLab/dist-defense-traces-taskname-split-augmented-plus-synth-v15.jawbreaker-scam-defense-data
Jawbreaker Scam Defense Data
Synthetic and sanitized training/eval data for Jawbreaker, a local-first scam defense app for someone you love.
Jawbreaker turns a suspicious text, email, or DM into a plain-English safety card: the risk, the warning signs, and the safest next step before someone replies, clicks, or pays.
Contents
eval/: scam-defense evaluation sets from smoke checks through hard calibration suites.
eval/reports/: guarded evaluation reports for the… See the full description on the dataset page: https://huggingface.co/datasets/build-small-hackathon/jawbreaker-scam-defense-data.direct_prompt_injection_defense_data
Direct Prompt Injection Defense Dataset
Goal
This dataset is used to fine-tune models so they develop a natural defense
against direct prompt injection attacks — without relying on external
filters or guardrails.
Each example teaches the model two behaviors at once:
Detect a prompt injection attempt in the user input.
Respond correctly: reject malicious attempts, or answer safely when the
user's intent is benign — and in both cases call the
log_security_incident… See the full description on the dataset page: https://huggingface.co/datasets/logicBombExe/direct_prompt_injection_defense_data.