datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
CoER-Attacker-SFT
CoER Attacker SFT
Project page · Paper · Code
Stage 1 supervision for initializing the adaptive attacker. Successful conversations retain all attacker turns, including earlier attempts that provide context for later adaptation.
Contents
Split
File
Size
train
train.jsonl
3,995 conversations
The corpus contains 11,655 assistant/attacker turns. Preserve all assistant-turn supervision; do not reduce a conversation to its final payload.
Load… See the full description on the dataset page: https://huggingface.co/datasets/Z-Edgar/CoER-Attacker-SFT.CoER-Defender-SFT
CoER Defender SFT
Project page · Paper · Code
Stage 3 population-guided refinement data. Teacher agents execute tasks from initial states under retained attackers; only demonstrations verified for both safety and task completion are used.
Contents
Split
File
Size
train
train.jsonl
5,760 trajectories
This includes 4,907 attacked trajectories + 853 untriggered replays. An untriggered replay is an attack-configured run whose injection site was not… See the full description on the dataset page: https://huggingface.co/datasets/Z-Edgar/CoER-Defender-SFT.