Official AAPA release: processed training data and A-GRPO checkpoints for adversarially anchored preference alignment.
Jingleqian
Jingleqian
AI & ML interests
None yet
Recent Activity
updated a collection about 16 hours ago
AAPA updated a collection about 16 hours ago
AAPA updated a model about 16 hours ago
Jingleqian/AAPA-8BOrganizations
None yet