arXiv 2508.15260

Deep Think with Confidence

By Yichao Fu, Xuewei Wang, et al.

Published 2025-08-21

Wiki summary

Explore the paper's summary, context, and related research on Papiers.

Large Language Models (LLMs) have shown great potential in reasoning tasks through test-time scaling methods like self-consistency with majority voting. However, this approach often leads to diminishing returns in accuracy and high computational overhead. To address these challenges, we introduce Deep Think with Confidence (DeepConf), a simple yet powerful method that enhances both reasoning efficiency and performan…

View the original paper on arXiv