研究人员推出了一种新颖的、无需训练的协作并行思维（CPT）框架，旨在提高大型语言模型测试时扩展（TTS）的效率。CPT通过实现不同分支之间的搜索时信息共享，解决了并行TTS方法中冗余探索的问题。这使得各分支能够重用他人已有的发现，而不是重新发现相同的信息，从而在HMMT和AIME等基准测试中改善了准确性-延迟权衡。 AI
影响 通过减少推理过程中的冗余计算，提高了大型语言模型的推理效率。
排序理由 该集群包含一篇详细介绍大型语言模型推理新方法的论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2605.27030v1 Announce Type: new Abstract: Test-Time Scaling (TTS) enhances the reasoning capabilities of large language models by allocating additional inference compute to explore the solution space. However, existing parallel TTS methods typically keep branches isolated d…
Test-Time Scaling (TTS) enhances the reasoning capabilities of large language models by allocating additional inference compute to explore the solution space. However, existing parallel TTS methods typically keep branches isolated during search: intermediate discoveries remain br…
Collaborative Parallel Thinking (CPT) enables information sharing across parallel search branches during inference to reduce redundant exploration and improve efficiency in test-time scaling for language models.