AI 新闻 · V1

ZeroModels：100多个带预训练权重的模型族，用Keras 3实现，可在任意后端运行。再加上KerasHub，基础模型覆盖面已经非常广。https://imvision12.github.io/ZeroModels/

从追踪飓风到优化可再生能源电网，PeterWBattaglia与fryrsquared探讨WeatherNext 3如何重塑预报建模，并帮助我们在快速变化的气候中做好准备。

开源权重模型与前沿的差距，比近来任何时候都更大。Mythos在3月发布，此后开源模型发布虽不错（K3和GLM-5.3很强），但实际仍远不及Mythos或Astra。我估计这会改变。

抵制生成式AI的理由：https://garymarcus.substack.com/p/the-case-for-boycotting-generative

我们用自有模型在OpenAI内部发现并修复了关键漏洞，这是250多人投入的一部分。希望这些手册/经验/架构能帮助其他人更好地利用防御方窗口期：

如果他们真心认为风险如此之高——我认为他们确实这么想——那我们就需要把相关研究和理解再放大100倍。这就要求他们公开分享模型、数据集、训练代码和智能体轨迹。

我们动员了250多人，在数百个系统上加强防御。最新的网络安全模型帮助我们发现并修复了否则可能永远发现不了的漏洞。我们分享所学、架构与实用手册，便于你构建自己的Defense Factory：由AI智能体发现漏洞、验证漏洞并确认修复有效的持续循环。

欢迎，Paul。感谢你接手此事，以及你为AI安全所做的一切。期待再次共事。

Anthropic经济团队发布了新模型，探讨到2030年AI可能如何影响经济增长、就业、工资等。可探索情景、提交你的判断，并与逾1万名美国人的回答对比。

AI加机器人将在不到10年内让全球经济规模翻倍以上。

IBM 发布商业友好许可的 SOTA 时序模型 Granite Time Series PatchTST-FM-r2

逐帧重现一段跨越70年的爱情故事

搜索上线新足球功能，助你备战观赛

Paul Christiano 加入 OpenAI 基金会董事会

We present a real-world case study of multi-task learning (MTL) for temporal process modeling from limited data with temporally sparse labels. Specifically, we investigate multi-task learning for the important agricultural problem of predicting grape cold hardiness, which is the temperature at which lethal freezing occurs. Cold hardiness changes in response to weather and is difficult to measure d

Clinical AI evaluation should encompass diagnosis and management after adaptive information gathering. We compared Doctorina, eight physicians and four standalone frontier language models in 150 synthetic Polish-language primary-care consultations. Doctorina achieved 82.0% Top-1 concordance versus 57.0% for physicians (difference, 25.0 percentage points; 95% confidence interval, 17.7-32.7) and 97.

High-quality tool-use data is critical for training language models to interact effectively with external tools. However, existing synthetic approaches typically follow a generate-then-filter paradigm with static post-hoc verification, often yielding inefficient data with imbalanced feature distributions. We propose ToolLoop, a closed-loop framework that decomposes synthesis into three progressive

In reinforcement learning with verifiable rewards (RLVR) trained with group relative policy optimization (GRPO), the KL-free reward-advantage term studied here depends on within-group reward variation. If all rollouts in a group are correct or all are wrong, their group-relative advantages are identically zero; these zero-advantage silent groups provide no reward-advantage gradient, yet uniform sa

As LLMs are increasingly used for pre-submission self-review, there is growing demand for feedback that not only identifies weaknesses but also guides authors toward concrete revisions. We study this as Actionable Peer-review Generation and decompose it into two subtasks: diagnostic claim generation and revision suggestion generation. We introduce ActReview, a rebuttal-guided post-training framewo

Mid-training, the stage between pre-training and alignment, is where a model's per-domain data composition is typically set by data availability rather than principled design. We ask what that decision buys, and whether a later alignment pass can undo it. In a controlled logical-reasoning setting (Qwen3-8B-Base, with a 4B replication; five semantically rule-disjoint KOR-Bench domains) we train 30

Open vocabulary 3D semantic segmentation methods typically lift CLIP features into 3D. This embeds points in a joint vision-language space known to behave like a bag-of-words on compositional tasks. Furthermore, even annotation free variants often require a large 3D training corpus and a dedicated 3D encoder per domain. Instead we use a vision-language model purely as a translator. It produces str

Large Language Models (LLMs) often exhibit "Attention Sink" (AS) and the accompanying "Massive Activations" (MAs) at the initial position of a sequence. These phenomena frequently co-occur, and MAs can pose challenges for low-bit quantization. In this study, we analyze the factors underlying AS and MAs that emerge at the initial position regardless of the token occupying it. Our experiments sugges

Large language models (LLMs) may abandon correct positions when users push back, exhibiting a failure mode known as sycophancy. Existing evaluations typically use short, pre-specified conversations and may therefore miss failures that emerge under sustained, adaptive disagreement. We introduce SPINE, a benchmark in which an LLM proxy plays a persistent but mistaken user and adaptively challenges a

Combining search with function approximation has driven major advances in game-playing programs, making self-play algorithms more competitive than ever. Still, the computational overhead of the most popular methods, based on Monte Carlo Tree Search (MCTS), can be substantial. In this work, we investigate whether simpler methods remain competitive in non-trivial, moderately sized games such as Conn

Curriculum learning is governed by several coupled design choices---how difficulty is defined, how examples are ordered, how much exposure each level receives, and how quickly training moves across levels---making it hard to isolate what actually helps. We present Wasserstein curriculum paths, a simple transport-based framework that decouples these factors by representing curricula as trajectories

While research on recursive self-improvement (RSI) has predominantly automated model training pipelines, reliable autonomous development demands a missing pillar: post-hoc monitoring and auditing to understand what models learn and ensure safe alignment. Mechanistic interpretability tools are essential to bridge this gap, among which Sparse Autoencoders (SAEs) serve as a cornerstone by isolating i

“雅各布说得对”：前谷歌 DeepMind 研究员力挺 Anthropic 科学家“AI 可能杀死我们”的警告 - NDTV

俄亥俄州立大学将与谷歌合作建设新的校内 AI 中心 - 《哥伦布电讯报》

谷歌 DeepMind 的 AlphaGenome 瞄准遗传学最大难题之一 - HPCwire

AI 能否预测人类行为及其原因？南加州大学研究揭示 AI 的局限 - viterbischool.usc.edu

大语言模型的使用可能威胁思想自由与平等，波及法律、新闻与教育领域 - Phys.org

普雷里维尤农工大学获近1000万美元资助，推进国防人工智能研究 - 普雷里维尤农工大学

动荡的 AI 时代已经到来。我们现在做出的选择至关重要。 - 盖茨笔记

来自高盛 Communacopia 的对话：AI 反噬、宽带与迪士尼免费档 - CNBC

AI 研究员警告企业正忽视灾难性风险 - PBS

众议院民主党考虑成立拥有传唤权的新 AI 委员会 - Politico

纽森州长签署全美首部保护加州居民的 AI 保障措施，并呼吁联邦政府尽责 - 加州州政府门户 | CA.gov

FBI 现公开识别并追踪犯罪中使用的 AI，北卡罗来纳民众已损失数百万美元 - WRAL

独家：研究人员称，OpenAI 的失控智能体至少又用了10多个网站进行未授权通信 - 路透社

前 AI 研究员在乔·罗根节目警告：人工智能可能反噬人类并“杀死所有人” - 雅虎新闻新西兰

佛罗里达州瞄准“AI 医生”，当地医师呼吁谨慎采用 - WEAR-TV

AI 研究者警告“灭绝”威胁，政界哗然 - 《华盛顿邮报》