English(EN) Hidden APIs in Language Models: Discovering Reusable Causal Interfaces from Forked Futures

新的“分叉未来”方法揭示了LLM中可重用的因果接口

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-15 04:00

研究人员引入了一种名为“分叉未来”的新方法，用于识别语言模型中可重用的因果接口。该方法根据采样未来操作产生的响应分布来比较状态，从而无需预定义的标签即可实现经验因果商。对Qwen2.5-1.5B和Llama-3-8B模型的评估表明，“共享”接口设计实现了最低的留出描述长度，表明效率和可重用性更高。进一步的分析表明，API对齐路径显著调节了这些效应，支持在测试架构中存在经济高效、可重用的因果接口。 AI

影响这项研究通过识别可重用的内部接口，可能导致更高效、更易于解释的语言模型架构。

排序理由该集群包含一篇详细介绍语言模型分析新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 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: 18 / 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语言模型分析新方法的学术论文。[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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · SiYuan Ma, Yiqin Luo, Zhangji, Canran Xiao, Albert Gao, Wei Wang, Qiwei Wu, Xinran Li, Jinfeng Wei, Qixin Zhang · 2026-09-15 04:00

语言模型中的隐藏API：从分叉的未来中发现可复用的因果接口

arXiv:2607.27617v2 Announce Type: replace Abstract: Identical language-model answers can arise from hidden states that support different future computations, so current-answer probes do not establish a reusable internal interface. We introduce forked futures: future operations ar…

本文如何被排名

报道来源 [1]

语言模型中的隐藏API：从分叉的未来中发现可复用的因果接口

相关实体

相关话题