English(EN) When to Ponder: Adaptive Compute Allocation for Code Generation via Test-Time Training

新的PonderTTT方法优化了代码生成的LLM计算

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-24 04:00

研究人员开发了PonderTTT，一种用于大型语言模型自适应计算分配的新颖策略，特别适用于代码生成任务。该方法利用测试时训练（TTT）层的自监督重构损失来动态决定何时应用更新，而无需单独的学习分类器。在The Stack v2数据集上使用GPT-2模型进行的实验表明，PonderTTT可以实现显著的性能提升，优于随机跳过，并且即使在分布外语言上也能保持高Oracle恢复率。 AI

影响该方法可能导致更高效的LLM推理，降低代码生成任务的计算成本。

排序理由该集群包含一篇详细介绍LLM计算优化新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍LLM计算优化新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 60 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Gihyeon Sim · 2026-08-24 04:00

何时思考：通过测试时训练实现代码生成的自适应计算分配

arXiv:2601.00894v2 Announce Type: replace-cross Abstract: Large language models apply uniform computation to all inputs, regardless of difficulty. We propose PonderTTT, a gating strategy using the TTT layer's self-supervised reconstruction loss to selectively trigger Test-Time Tr…

本文如何被排名

报道来源 [1]

何时思考：通过测试时训练实现代码生成的自适应计算分配

相关实体

相关话题