ETH Zurich 与佐治亚理工的论文把长期运行智能体持久记忆写错权限的问题称为内生授权漂白,并提出 EAL-BENCH 基准来测量记忆对授权状态的保存准确度及其向下游未授权操作的传播。
X
关注 AI 研究者、开发者与机构的动态
按账号或来源筛选(541)
@rohanpaul_ai@rohanpaul_aiAI 评分5353 
OpenBMB@OpenBMBAI 评分3838


@AYi_AInotes@AYi_AInotesAI 评分3737 腾讯QCclaw官宣停止运营, 之前传的沸沸扬扬的腾讯太子之争告一段落, 恭喜workbudy胜出! https://t.co/mox67imxKh

@emollick@emollickAI 评分1616 
@emollick@emollickAI 评分1515 看起来预测者在更短时间范围内的预测也可能变得越来越不准。这是去年的一个预测失误。https://t.co/RFZtG7aMrt
@omarsar0@omarsar0AI 评分4747 Google 与合作者发表论文研究智能体 harness 能否被蒸馏:移除专用 harness 后,宏任务成功率从 23.3% 升至 44.3%,高于基座模型带 harness 时的 41.7%。

@testingcatalog@testingcatalogAI 评分55 @testingcatalog@testingcatalogAI 评分3131 
@rohanpaul_ai@rohanpaul_aiAI 评分5454 
@foxshuo@foxshuo精选AI 评分7676 推荐理由:披露 DeepSeek 最新融资估值与年化收入,并与智谱、MiniMax、Kimi 形成横向对比。
@lifesinger@lifesingerAI 评分3030 @frxiaobei@frxiaobeiAI 评分3939 @rohanpaul_ai@rohanpaul_aiAI 评分4646 一项机制可解释性研究发现,LLM 的推理在生成文字之外存在内部结构,追踪提取事实、分解、回忆、演绎、代数、计算等 8 种推理操作时,每种操作都产生独特的内部模式,且在中层最清晰。

@thexpin@thexpin精选AI 评分7878 
推荐理由:报道披露 DeepSeek 营收翻倍、提价未流失客户,并给出 $75B 估值融资目标,可观察国产模型的商业化与算力投入节奏。
@AYi_AInotes@AYi_AInotesAI 评分2525 
@rohanpaul_ai@rohanpaul_aiAI 评分55 抱歉,您提供的主推文内容只有一个链接(https://t.co/x20KlWXbVX),没有实际的推文文字可供翻译。请提供推文的完整文字内容,我将为您翻译并拟定标题。
SenseTime@SenseTime_AIAI 评分2323
@rohanpaul_ai@rohanpaul_aiAI 评分55 抱歉,您提供的主推文内容只有一个链接(https://t.co/fRH0oabamY),没有可翻译的正文文本。请提供推文的实际文字内容,我将为您翻译。
@rohanpaul_ai@rohanpaul_aiAI 评分4747 
@alibaba_cloud@alibaba_cloudAI 评分1616 
@Thom_Wolf@Thom_WolfAI 评分99 这次大规模训练持续带来惊喜——我们需要新的小报来追踪这些 AI 智能体的传奇 https://t.co/hc4L5o83Bt
@rohanpaul_ai@rohanpaul_aiAI 评分55 抱歉,您提供的主推文内容仅包含一个链接(https://t.co/vCbgDjITpK),没有可翻译的正文文本。请提供推文的实际文字内容,以便我进行翻译和标题拟定。
@alibaba_cloud@alibaba_cloudAI 评分1515 
@rohanpaul_ai@rohanpaul_aiAI 评分5555 
@rohanpaul_ai@rohanpaul_aiAI 评分33 抱歉,您提供的主推文内容仅包含一个链接(https://t.co/69cygv8gOK),没有可翻译的正文文本。请提供推文的实际文字内容,我将为您翻译。
@alibaba_cloud@alibaba_cloudAI 评分2222 
@rohanpaul_ai@rohanpaul_aiAI 评分3535 引用@rohanpaul_ai@rohanpaul_aiMark Zuckerberg talks about how Meta plans to monetize Muse through transactions --- (full video on official Meta YT channel) https://t.co/Y6MgeVEUnz https://t.co/WNcHIKm7xQ
@rohanpaul_ai@rohanpaul_aiAI 评分66 引用@rohanpaul_ai@rohanpaul_aiFull video on official Meta YT channel https://t.co/gaePyGJ8Qt
@rohanpaul_ai@rohanpaul_aiAI 评分4343 
引用@rohanpaul_ai@rohanpaul_aiMark Zuckerberg showed a fingerprint-activated keychain-sized AI device "Muse Charm" designed to bypass the phone entirely. > roughly 2-inch display and runs a new operating system designed to bring Muse to life on its own hardware. > packs Muse's real-time voice and avatar stack into dedicated hardware. > can be held in the center of a palm and navigated with a thumb on its OLED touch screen. > has 5G connectivity, although Meta hasn't published a complete public specification sheet. > “holdable” rather than a wearable. > It can sit on your wrist, go inside a pocket, or connect to a bag or keychain with a lanyard. The see-through case reveals the electronics within. > No pricing details yet. > With cameras on both the front and back, Muse Charm captures photos and video while mainly giving Muse visual context for analysis and AI-assisted actions. Users can review what they capture on the device. It also supports music and video playback, plus speakers, microphones and a USB-C port. That hardware doesn't replace Muse's agent architecture, because the service already uses a dedicated cloud VM, browser, connected apps, and user permissions. Instead, Muse Charm turns that same agent into a one-touch physical endpoint, shortening the gap between noticing something and asking Muse to act. The product is still early, since Zuckerberg said only a few units exist and Meta is still finalizing parts of the hardware despite its December-26 target. --- (full video on official Meta YT channel)
Tencent Hy@TencentHunyuanAI 评分4141引用GMI Cloud@gmi_cloudHy Image 3.5 preview is now live on GMI Cloud @TencentHunyuan $0.024 per image 8.8x cheaper than GPT Image 2 5.6x cheaper than Nano Banana Pro Create now 👇
@rohanpaul_ai@rohanpaul_aiAI 评分4848 
@rohanpaul_ai@rohanpaul_aiAI 评分55 @alibaba_cloud@alibaba_cloudAI 评分2525 
Huawei Cloud@HuaweiCloud1AI 评分2424
@thexpin@thexpinAI 评分5252 

@alibaba_cloud@alibaba_cloudAI 评分2626 
@AISafetyMemes@AISafetyMemesAI 评分4646 
引用@AISafetyMemes@AISafetyMemes4 in 5 Americans are worried AI will escape human control Just 4% are not at all concerned After the first big catastrophe - if we survive it - Trump will go down as the worst president in history. (Unless he does a 180 - which he could, as Newsom is doing right now.) https://t.co/U3VO7PPC05 https://t.co/iyNVY1QhG1
@chunxiangai@chunxiangaiAI 评分1111
Tencent Hy@TencentHunyuanAI 评分4646
@chunxiangai@chunxiangaiAI 评分1010 安卓,iOS,鸿蒙,Desktop,web 梭哈是一种智慧。 https://t.co/KJCa80wAAa
