Understanding Inference Scaling for LLMS: Bottlenecks, Trade-Offs, and Performance Principles.
Moiz Arif, Avinash Maurya, Sudharshan Vazhkudai, Bogdan Nicolae
Browse the full ISCA paper archive.
Moiz Arif, Avinash Maurya, Sudharshan Vazhkudai, Bogdan Nicolae
Browse the full ISCA paper archive.