3 papers
cs.CL2026
KALAVAI: Predicting When Independent Specialist Fusion Works -- A Quantitative Model for Post-Hoc Cooperative LLM Training
Ramchand Kumaresan
Independently trained domain specialists can be fused post-hoc into a single model that outperforms any individual specialist, and the gain is predictable: gain = 0.82 x divergence…
cs.LG2026
Orion: Characterizing and Programming Apple's Neural Engine for LLM Training and Inference
Ramchand Kumaresan
Over two billion Apple devices ship with a Neural Processing Unit (NPU) - the Apple Neural Engine (ANE) - yet this accelerator remains largely unused for large language model workl…
cs.LG2026
ACAR: Adaptive Complexity Routing for Multi-Model Ensembles with Auditable Decision Traces
Ramchand Kumaresan
We present ACAR (Adaptive Complexity and Attribution Routing), a measurement framework for studying multi-model orchestration under auditable conditions. ACAR uses self-consistency…