通过温度调控提升大模型推理能力，让思考更高效准确。
Guidance is All You Need: Temperature-Guided Reasoning in Large Language Models
- 用热/冷令牌机制动态调整词汇重要性，引导推理路径
- 在多个任务上推理准确率显著提升，计算效率更高
- 适合需要高质量逻辑推理的场景，如数学题与复杂决策
我们提出 Quasar-1，一种新架构，通过令牌温度机制（TTM）和引导思维链（GSoT）引入温度引导推理。该方法基于热令牌与冷令牌概念：热令牌优先体现上下文相关性，冷令牌提供补充信息。这种动态调节令牌重要性的机制，使模型在逻辑推理方面优于传统思维链方法。通过严谨的数学分析，我们证明温度引导注意力机制能以指数级保证收敛到最优推理路径。实证结果表明，在多种任务中，该方法显著提升了推理准确率与计算效率，推动高级人工智能推理向更广泛应用场景普及。
原文摘要 · Abstract (English)
We present Quasar-1, a novel architecture that introduces temperature-guided reasoning to large language models through the Token Temperature Mechanism (TTM) and Guided Sequence of Thought (GSoT). Our approach leverages the concept of hot and cold tokens, where hot tokens are prioritized for their contextual relevance, while cold tokens provide supplementary information. This dynamic modulation of token importance enables the model to achieve superior logical reasoning capabilities compared to traditional chain-of-thought approaches. Through rigorous mathematical analysis, we prove that our temperature-guided attention mechanism converges to optimal reasoning paths with exponential guarantees. Empirical results show significant improvements in reasoning accuracy and computational efficiency across a wide range of tasks, making advanced AI reasoning accessible to a broader range of applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。