English(EN) InterPol: De-anonymizing LM Arena via Interpolated Preference Learning

新框架InterPol可对LM Arena响应进行去匿名化

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-31 04:00

一篇新的研究论文介绍了一个名为InterPol的框架，该框架旨在对LM Arena等排行榜上的语言模型响应进行去匿名化。与依赖简单统计特征的先前方法不同，InterPol合成了插值偏好数据，以捕捉更深层次的风格模式。实验表明，InterPol在识别准确性方面显著优于现有基线，并通过排名操纵模拟展示了其现实世界的威胁。 AI

影响凸显了LLM排行榜潜在的漏洞以及对更强大匿名措施的需求。

排序理由研究论文，详细介绍了一种对语言模型响应进行去匿名化的新方法。[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: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 研究论文，详细介绍了一种对语言模型响应进行去匿名化的新方法。[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, safety; 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: 3 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.AI TIER_1 English(EN) · Minsung Cho, Jaehyung Kim · 2026-10-31 04:00

InterPol：通过插值偏好学习对LM Arena进行去匿名化

arXiv:2603.15220v3 Announce Type: replace Abstract: Strict anonymity of model responses is a key for the reliability of voting-based leaderboards, such as LM Arena. While prior studies have attempted to compromise this assumption using simple statistical features like TF-IDF or b…

本文如何被排名

报道来源 [1]

InterPol：通过插值偏好学习对LM Arena进行去匿名化

相关实体

相关话题