Active Imitation Learning for Thermal- and Kernel-Aware LFM Inference on 3D S-NUCA Many-Cores

Develop a lightweight scheduling library for LFM inference on many-core systems that accounts for thermal and cache locality. This is critical for deploying local models on hardware other than traditional GPUs.

Suggested repo: ThermalCore

"Optimized LFM inference that keeps your cores cool."

Estimated effort: 90h