Saved in:
Bibliographic Details
Main Authors: Zhao, Youpeng, LV, Jinpeng, Wu, Di, Wang, Jun, Gooley, Christopher
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.19645
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Test-time scaling (TTS) has recently emerged as a promising direction to exploit the hidden reasoning capabilities of pre-trained large language models (LLMs). However, existing scaling methods narrowly focus on the compute-optimal Pareto-frontier, ignoring the simple fact that compute-optimal is not always system-optimal. In this work, we propose a system-driven perspective on TTS, analyzing how reasoning models scale against practical metrics, such as latency and cost-per-token. By evaluating the impact of popular optimizations such as tensor parallelism and speculative decoding, our preliminary analysis reveals the limitations of current methods and calls for a paradigm shift toward holistic, system-aware evaluations that capture the true essence of scaling laws at inference time.