Explorar

for the last time - SEO is NOT dead! it has changed with AI and this is why I'm building Outrank - to help you get recommended by ChatGPT & rank on Google

Built Tweet Hunter, Taplio (sold $8m) Growing https://t.co/OyNJ8ZUyOh - https://t.co/jS9GQJ5Ps8 - https://t.co/EFUcKeBbpU - https://t.co/JkVOl1O0S1 - https://t.co/KG9PgxJabg Sharing weekly tips about growth: https://t.co/ereQodN3Ov

Tibo

Tue Nov 11 12:14:45

RT @shao__meng: 「Google x Kaggle 5天 AI 智能体强化课程」第二天，Google 11月最新重磅白皮书「Agent Tools & Interoperability with MCP」这份白皮书，探讨了基础模型如何通过外部工具扩展能力，重点…

邵猛，中年失业程序员 😂 专注 - Context Engineering, AI Agents. 分享 - AI papers, apps and OSS. ex Microsoft MVP 合作 - 私信/邮箱：shaomeng@outlook.com 📢 公众号/小红书: AI 启蒙小伙伴

meng shao

Tue Nov 11 12:14:15

Bullish. In terms of raw capacity it's still ≈V3-tier. This goes to show that LisanBench is indexing benefits of well-done RL, and Moonshot's RL is quite good. Grok-fast is surprising.

We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1

Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)

Tue Nov 11 12:13:42

Within the current AI paradigm (deep learning with SGD), unlearning (forgetting) is technically near-impossible without full retraining or with formal guarantees. Once learned, we persist... not as data, but as gradients fossilized in the model’s weights.

Exploring science, cognition, AI, robotics, engineering, dynamical complex systems, more importantly life and beyond. Founder at https://t.co/HInAbA9KEW, and https://t.co/vnHrcX97UC

Faruk “Frank” Guney

Tue Nov 11 12:13:40

每天玩 AI 工具还能拿工资？没错，这正是全球阅读量最大的每日 AI 新闻通讯 @TheRundownAI 在招的「Copywriter - AI」职位，看看这个职位要做什么，有哪些要求，对于 AI 创作者也会有所启发。岗位职责 · AI 工具评估：测试所有新发布的 AI 工具，分析其实际价值，并解释为什么某些工具值得采用、哪些则不宜使用。 · 内容创作：撰写实用指南和文章，帮助用户高效运用 AI 工具。每周需输出5篇内容，触达数百万读者。 · 动态追踪：充当“AI 咨询专家”，通过 X 等平台提前捕捉最新趋势，提供客观洞见。任职要求 · 核心技能：具备文案写作、新闻报道或内容创作经验（AI 相关背景优先）；出色的研究能力，能在 X 等平台领先一步把握AI动态；强烈的 AI 直觉——辨识何时依赖 AI、何时需人工干预。 · 工作风格：每周5篇输出节奏，远程灵活，但需保持高产与高质量。 · 其他：无特定学历门槛，但需展示对 AI 的深刻理解和内容影响力。

邵猛，中年失业程序员 😂 专注 - Context Engineering, AI Agents. 分享 - AI papers, apps and OSS. ex Microsoft MVP 合作 - 私信/邮箱：shaomeng@outlook.com 📢 公众号/小红书: AI 启蒙小伙伴

meng shao

Tue Nov 11 12:13:38

还有，一张在深圳北拍摄的夜景图（图 1）。不给任何提示，它能根据图中的建筑特征，判断出具体的拍摄位置（图 2）。而且是有理有据的，真的有点厉害。

总的来看，这次模型引入的 “图像思考” 能力，确实挺强的，似乎真的能看懂并理解图片信息。目前还没有在线体验地址，不过模型已经开源。模型下载：https://t.co/WOYawrwuZL

GitHubDaily

Tue Nov 11 12:13:34

Newest first — browse tweet threads

Explorar

Newest first — browse tweet threads

for the last time - SEO is NOT dead! it has changed with AI and this is why I'm building Outrank - to help you get recommended by ChatGPT & rank on Google

RT @shao__meng: 「Google x Kaggle 5天 AI 智能体强化课程」第二天，Google 11月最新重磅白皮书「Agent Tools & Interoperability with MCP」这份白皮书，探讨了基础模型如何通过外部工具扩展能力，重点…

Bullish. In terms of raw capacity it's still ≈V3-tier. This goes to show that LisanBench is indexing benefits of well-done RL, and Moonshot's RL is quite good. Grok-fast is surprising.

Within the current AI paradigm (deep learning with SGD), unlearning (forgetting) is technically near-impossible without full retraining or with formal guarantees. Once learned, we persist... not as data, but as gradients fossilized in the model’s weights.

还有，一张在深圳北拍摄的夜景图（图 1）。不给任何提示，它能根据图中的建筑特征，判断出具体的拍摄位置（图 2）。而且是有理有据的，真的有点厉害。

Explorar

Newest first — browse tweet threads

Explorar

Newest first — browse tweet threads

for the last time - SEO is NOT dead! it has changed with AI and this is why I'm building Outrank - to help you get recommended by ChatGPT & rank on Google

RT @shao__meng: 「Google x Kaggle 5天 AI 智能体强化课程」第二天，Google 11月最新重磅白皮书「Agent Tools & Interoperability with MCP」 这份白皮书，探讨了基础模型如何通过外部工具扩展能力，重点…

Bullish. In terms of raw capacity it's still ≈V3-tier. This goes to show that LisanBench is indexing benefits of well-done RL, and Moonshot's RL is quite good. Grok-fast is surprising.

Within the current AI paradigm (deep learning with SGD), unlearning (forgetting) is technically near-impossible without full retraining or with formal guarantees. Once learned, we persist... not as data, but as gradients fossilized in the model’s weights.

还有，一张在深圳北拍摄的夜景图（图 1）。 不给任何提示，它能根据图中的建筑特征，判断出具体的拍摄位置（图 2）。 而且是有理有据的，真的有点厉害。

RT @shao__meng: 「Google x Kaggle 5天 AI 智能体强化课程」第二天，Google 11月最新重磅白皮书「Agent Tools & Interoperability with MCP」这份白皮书，探讨了基础模型如何通过外部工具扩展能力，重点…

还有，一张在深圳北拍摄的夜景图（图 1）。不给任何提示，它能根据图中的建筑特征，判断出具体的拍摄位置（图 2）。而且是有理有据的，真的有点厉害。