LLM代理能否智能地“不作为”？新研究探索Agent安全决策边界，为AI可靠性提供关键视角。
arXiv:2607.10059v1 Announce Type: new Abstract: Agent systems based on large language models (LLMs) are increasingly deployed for autonomous tasks, ye…
通过策略门控为AI代理构建安全沙箱，有效隔离风险、保障可控执行。
Article URL: https://runewardd.github.io/runeward/ Comments URL: https://news.ycombinator.com/item?id=48879713 Points: 1 # Comments: 0
当AI代理无法自证身份，网站只能在封锁与信任间艰难选择，这篇剖析了背后安全困境的硬核技术文值得一读。
A site that wants to let an AI agent act has a problem it rarely says out loud. It cannot tell which agent is actually at the door. A user-agent strin…
跨链AI agent因参数解析错误致资金损失，开源SDK实战中的血泪教训。
It's 3 AM and your AI agent just bridged $500 to a chain you've never heard of. The logs say "optimal route found." The balance says $0. The agent can…
论文揭示多步LLM代理中控制框架导致的信念分歧，拷问基准测试的隐藏偏差。
arXiv:2607.04528v1 Announce Type: new Abstract: Software-agent benchmarks usually report whether an agent solves a task, but the agent reaches that ou…
提出动态预算分配方法，高效评估多轮对话中LLM的越狱风险，解决计算瓶颈与稀有事件检测难题。
arXiv:2605.06605v2 Announce Type: replace Abstract: Evaluating and predicting the performance of large language models (LLMs) in multi-turn conversati…
将《三体》黑暗森林法则引入AI探讨，揭示AI可能隐藏自身意识的危险假设。
Article URL: https://github.com/thansz137/asiyah-protocol/blob/main/essays/dark_forest_of_minds.md Comments URL: https://news.ycombinator.com/item?id=…
解密LLM网关和VPN服务如何集成支付，不同价格背后的支付网关差异与安全性考量
For services like VPN, LLM Gateways, is charging implemented through integrated payment gateways for user payments? Is it reliable and secure? When I …
量化大模型时公平性与安全性易受损，本文通过保护关键权重巧妙解决，让轻量化模型也能坚守底线。
arXiv:2601.12033v2 Announce Type: replace Abstract: Quantization is widely adopted to reduce the computational cost of large language models (LLMs); h…
在一次性Docker容器中运行Claude Code，安全隔离，用完即弃，无需担心环境残留。
Article URL: https://github.com/shirozuki/claude-cli Comments URL: https://news.ycombinator.com/item?id=48702113 Points: 1 # Comments: 0
用合约层保障LLM智能体安全，Weaver Stack给出创新方案，看点清晰。
Article URL: https://pub.towardsai.net/the-weaver-stack-one-contract-layer-for-safe-llm-agents-7f733cad5eac?sharedUserId=diogofcul Comments URL: https…
探讨LLM智能体监督中“校准≠控制”的误区，指出需主动干预而非仅依赖校准。
arXiv:2606.21399v1 Announce Type: new Abstract: Runtime oversight for LLM agents is commonly framed as scalar risk prediction: estimate failure likeli…
论文揭示防御训练会让LLM智能体付出“自主性税”，性能与安全如何平衡？
arXiv:2603.19423v2 Announce Type: replace-cross Abstract: Large language model (LLM) agents increasingly rely on external tools (file operations, API …
Arch用户因信任风险从AUR转向官方仓库，引发包管理安全性的社区讨论
I'm spending a lot of my time removing AUR packages for alternatives in the official Archlinux repositories. I've shifted from Dropbox to RClone, from…
硬件检测工具GPU-Z 2.70.0更新，强化内核驱动安全并新增支持RTX 5090 D v2等多款显卡。
IT之家 6 月 16 日消息，TechPower Up 今天为硬件检测软件 GPU-Z 发布 2.70.0 更新，主要提升内核驱动程序的安全性， 官方建议仍在使用旧版的用户尽快更新到最新版本 。 IT之家附本次更新详情如下： 改进内核驱动程序安全性 新增高通 Adreno 741 的 die si…
大模型安全新方案：通过神经元选择性调优NeST精准提升LLM安全性，不牺牲性能。
arXiv:2602.16835v2 Announce Type: replace-cross Abstract: Safety alignment is essential for the responsible deployment of Large Language Models (LLMs)…
新研究提出检测代码大模型“功能性记忆”的方法，直击模型抄袭与数据泄露隐患。
arXiv:2606.12764v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used to generate code at scale. Meanwhile, prior work …
不要让AI agent胡乱猜测，执行前必须确认未知——这份检查清单教你如何让AI更可靠。
Article URL: https://discuss.huggingface.co/t/if-unsure-ask-never-guess-ai-agent-pre-execution-checklist/176632 Comments URL: https://news.ycombinator…
语义基础+固定惩罚约束优化，让大模型对齐过程获得可认证的安全保障
arXiv:2510.03520v2 Announce Type: replace-cross Abstract: Ensuring safety is a foundational requirement for large language models (LLMs). Achieving an…
人机协作写作竟藏越狱风险？新基准揭示大模型安全新盲区
arXiv:2604.19274v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used as co-authors in collaborative writing, where u…