Beyond Model Optimization: Practical Energy-Efficient LLM Inference Through Context-Aware Input Reduction.

Kalle Pronk, Qin Zhao, Siebren Kazemier

Browse the full SANER paper archive.