智能AI None 从无到有：语言模型能否发现 0？ 17289v1 Announce Type: new Abstract: AI systems based on artificial neural networks are being developed with aspirations that language models 2026-06-17 Phoebe Zeng, Thomas L. Griffiths, Brenden M. Lake
智能AI None 嵌入模型路由的政策遗憾：低级别专家的上下文强盗 14929v1 Announce Type: new Abstract: Modern recommendation systems increasingly rely on dynamically routing diverse quer and the models 2026-06-16 Yan Dai, Negin Golrezaei, Patrick Jaillet
智能AI None GRASP：梯度对齐顺序参数传输，实现内存高效的多源学习 arXiv:2606.14900v1 Announce Type: new Abstract: Multi-source transfer learning faces a fundamental scalability bottlenec source memory and 2026-06-16 Mary Isabelle Wisell, Nicholas Jacobs, Aayush Manandhar, Salimeh Yasaei Sekeh
智能AI None 关系结构因果模型 arXiv:2606.14892v1 Announce Type: new Abstract: An artificial intelligence must have a model of its environment that is causal and relational 2026-06-16 Adiba Ejaz, Elias Bareinboim
智能AI None 重温 WorkBench：工作场所代理两年后 13715v1 Announce Type: new Abstract: The best agent on WorkBench in March 2024, GPT-4, completed 43% of tasks and took a the and agent 2026-06-15 Olly Styles
开发者生态 None 在家进行人工智能编码而不会破产 There are three ways to do AI coding at home without spending like a company, and which one fits depends mostly on how m the and you 2026-06-13 sbochins
安全攻防 None 美国命令 Anthropic 暂停外国人观看《神鬼寓言 5》和《神话 5》 Anthropic said on Friday it will "abruptly disable" its most advanced artificial intelligence (AI) models, Claude Fable the said models 2026-06-13 info@thehackernews.com (The Hacker News)
智能AI None 用于半导体制造的基于物理的生成人工智能：通过构建在生成模型中实施严格的物理约束 arXiv:2606.11247v1 Announce Type: new Abstract: Generative models are increasingly used to propose designs, data, and co and physical physics 2026-06-11 Yaser Mike Banad, Sarah Sharif
智能AI None 是否干预：通过概率模型混合指导推理时间对齐 11201v1 Announce Type: new Abstract: The wide deployment of LLMs has made model alignment necessary to make newly traine and alignment models 2026-06-11 Jin Gan, Xin Li, Jun Luo
智能AI None 通过 Oracle 云承诺访问 OpenAI 模型和 Codex Access OpenAI models and Codex through Oracle Cloud, using existing commitments to build and deploy AI with enterprise s and Access OpenAI 2026-06-11 OpenAI博客
智能AI None 模糊窗口注意 arXiv:2606.09862v1 Announce Type: new Abstract: The Softmax Attention operation in Transformer language models has a qua the and Attention 2026-06-10 Axel Laborieux, Christos Sourmpis, Juan Gabriel Kostelec, Qinghai Guo
智能AI None Elmes*：长尾教育场景中大型语言模型细粒度评估标准的自动构建 06546v1 Announce Type: new Abstract: Evaluating large language models (LLMs) for education requires measuring how models and for that 2026-06-08 Tao Liu, Ye Lu, Ruohua Zhang, Siyu Song, Wentao Liu, Aimin Zhou, Hao Hao
开发者生态 None 沉默的批评家 Like most folks, I’ve been using The Models 1 to write code now for the better part of a year。 the and that 2026-05-29 jfb
智能AI None 克劳德的新模型在混乱时更加“诚实” Anthropic is releasing Claude Opus 4。8 on Thursday, and the company is touting the model's "honesty。 that Anthropic the 2026-05-28 Jay Peters
开源推荐 None GitHub 热门项目：GLM-4 GitHub项目：GLM-4 仓库地址：https://github。 GLM and the 2026-05-28 GitHub Trending
智能AI None 约束获取需要更好的基准 arXiv:2605.26279v1 Announce Type: new Abstract: Constraint Acquisition (CA) and related research on the validation and e and the models 2026-05-27 Rafa{\l} Stachowiak, Tomasz P. Pawlak
智能AI None LLM可以反省吗？现实检验 26242v1 Announce Type: new Abstract: Can large language models detect and report their own internal states。A number of s the that their 2026-05-27 Shashwat Singh, Tal Linzen, Shauli Ravfogel
开发者生态 None 与前沿实验室相比，外包加上 LocalAI 很快就会变得更加经济 【HN用户评论摘要】 The current closed source frontier models are more capable than the latest from DeepSeek。But is the capabilit the are frontier 2026-05-26 GodelNumbering
智能AI None 多少思考才足够？量化和理解法学硕士推理中的冗余 23926v1 Announce Type: new Abstract: Reasoning-capable large language models solve hard problems by emitting long chains the and models 2026-05-26 Zhiyuan Zhai, Xinkai You, Wenjing Yan, Xin Wang
开源推荐 None GitHub 热门项目：大型语言模型实践 GitHub项目：Hands-On-Large-Language-Models 仓库地址：https://github。 Large Language Models 2026-05-25 GitHub Trending