PulseAugur
中
实时 03:36:24
English(EN) TRACE: A Two-Channel Robust Attribution Watermark via Complementary Embeddings for LLM-Agent Trajectories

新的TRACE水印确保LLM代理轨迹的来源

研究人员开发了TRACE,一种新颖的双通道水印,旨在确保LLM代理轨迹的来源。该系统能够抵御可能试图重新品牌化或替换代理的对手,因为它将水印直接嵌入到轨迹日志中。TRACE利用一个基于局部内容进行无失真动作选择的选择通道,以及一个基于日志骨架以在重写下保持不变的计数通道,从而确保即使日志被篡改也能进行归因。 AI

影响 该水印系统可以增强LLM代理部署中的信任和问责制,特别是在涉及第三方经销商的情况下。

排序理由 该集群描述了一篇详细介绍LLM代理轨迹水印的新颖技术方法的学术论文。

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

新的TRACE水印确保LLM代理轨迹的来源

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
Research
该集群描述了一篇详细介绍LLM代理轨迹水印的新颖技术方法的学术论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
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
101 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 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Zheng Gao, Xiaoyu Li, Xiaoyan Feng, Jiaojiao Jiang, Yang Song, Yulei Sui, Zhenchang Xing, Liming Zhu ·

    TRACE:一种通过互补嵌入实现大语言模型-代理轨迹的双通道鲁棒归因水印

    arXiv:2607.08400v1 Announce Type: cross Abstract: LLM agents reach users through resellers, who may rebrand a developer's agent or substitute a cheaper model. When provenance is disputed, attribution rests on the trajectory log (the record of tool calls, observations, and execute…

  2. arXiv cs.AI TIER_1 English(EN) · Liming Zhu ·

    TRACE:一种通过互补嵌入实现大语言模型-代理轨迹的双通道鲁棒归因水印

    LLM agents reach users through resellers, who may rebrand a developer's agent or substitute a cheaper model. When provenance is disputed, attribution rests on the trajectory log (the record of tool calls, observations, and executed actions, not the model's reasoning), which the r…