PulseAugur
中
实时 06:56:34
English(EN) From Bytes to Behavior: Request-Level Attribution from Live GPU Hidden States

新方法从 GPU 内存中提取实时 AI 模型行为

研究人员开发了一种新颖的方法,可以在不修改正在运行的 AI 系统的情况下,从实时 GPU 进程中提取中间隐藏状态。该技术通过直接从视频内存中捕获每层激活来分析已部署的大型语言模型的行为。该方法已在众多开源模型上得到验证,可用于下游分析,例如以高精度区分合规响应和拒绝响应。 AI

影响 实现了在不直接访问模型内部的情况下审计和理解已部署 AI 系统的新方法。

排序理由 详细介绍 AI 系统分析新技术的学术论文。[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 →

新方法从 GPU 内存中提取实时 AI 模型行为

本文如何被排名

Signal score
37 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
详细介绍 AI 系统分析新技术的学术论文。[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, infra
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]

  1. arXiv cs.AI TIER_1 English(EN) · Jonathan Pan, Yidi Yuan, Xin Wei Chia, Xinhui Su, Vincent Tan, Brenda Ong ·

    从字节到行为:来自实时 GPU 隐藏状态的请求级归因

    arXiv:2611.18272v1 Announce Type: cross Abstract: Existing memory forensics for machine learning systems typically targets static artifacts, such as model weights or prompt logs, rather than the transient computational states of a live inference process. This gap limits an examin…