Problem
Research questions and friction points this paper is trying to address.
test-time scaling
candidate generation
energy consumption
performance
large language models
Sample Count Is Not Enough: Candidate-Generation Strategy Shapes the Energy and Performance of LLM Test-Time Scaling
๐ค AI Summary
็ ็ฉถ้่ฟ่ฐๆด็ๆๅ้ๅๅบ็็ญ็ฅ๏ผไฝฟ็จไธๅๆนๆฌกๅคงๅฐ็ๆนๆณๆฅไผๅๅคงๅ่ฏญ่จๆจกๅๆจ็ๆถ็่ฝ้ๅๆง่ฝๆถ่้ฎ้ขใ
๐ Abstract
Test-time scaling can improve large language model reasoning by generating and combining multiple candidate responses. In sampling-based methods, the inference budget is often described by the number of generated candidates, N. However, N tells us how many candidates are generated, not how they are executed. The same candidate budget can be produced in one batched generation call or split across several sequential calls with smaller batch sizes. We first study the effect of increasing N on reasoning accuracy using Phi-3-mini and Qwen2.5-1.5B on 500 GSM8K prompts. As expected, increasing N from 1 to 8 improves accuracy by 8.4 percentage points for Phi-3-mini and 18.4 points for Qwen2.5-1.5B. However, accuracy alone does not show the systems cost of using a larger candidate budget. We therefore fix N = 8 and compare four generation schedules: 1x8, 2x4, 4x2, and 8x1, where axb denotes a generation calls with b candidates per call. We measure latency, throughput, GPU-hours, and gross GPU-device energy while keeping the total candidate count fixed. On A100 GPUs, eight serial calls use 4.64-4.86x as much gross GPU-device energy and have 5.77-6.12x the P95 latency of one batched call with eight candidates. The same pattern appears across three independently scheduled A100 nodes per model and in short-output SciQ/V100 experiments. These results show that candidate count alone is not enough to describe the systems cost of multi-candidate test-time scaling. When candidates are independent and memory allows it, fewer generation calls with larger batch sizes are more efficient. Evaluations should therefore report not only candidate count and accuracy, but also generation schedule and GPU-level systems metrics.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Test-time Scaling
Candidate-Generation Strategy
Energy Efficiency
Latency
Batch Size
๐ Similar Papers
No similar papers found.
