DeEscalWild: A Real-World Benchmark for Automated De-Escalation Training with SLMs

Create an end-to-end framework to fine-tune SLMs for high-stakes de-escalation scenarios. Focus on latency-optimized inference for deployment on edge devices.

Suggested repo: slm-deescalate

"Train small, act big: safe de-escalation for local edge hardware."

Estimated effort: 120h