Step-level reasoning monitors for LLM agents: StepSafe-4B, StepSafe-8B and the StepSafe-4B-SFT warm start.
Carrie Gu
Caaaarr1e
·
AI & ML interests
None yet
Recent Activity
upvoted a paper 8 days ago
StepSafe: Learning Timely Intervention with Context-Aware Safety Reasoning liked a model 11 days ago
Caaaarr1e/StepSafe-4B updated a collection 14 days ago
StepSafeOrganizations
None yet