- 论文公开站arXiv
Agent in a Bottle: Can LLM Agents Turn Their Capabilities Into Cheap, Scalable Artifacts?
Large language models (LLMs) can solve many narrow tasks, but querying them separately for millions of related instances can be prohibitively expensive. Can LLM agents autonomously create cheaper solutions for such workl…
- 资讯公开站rss_arxiv_cs_ai
Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas
arXiv:2610.03998v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents. Their validity as substitutes for particular respondents depends …
- 论文公开站arXiv
UniSlider: Perceptually Uniform Sliders for Continuous Image Editing
Sliders provide an intuitive interface for continuous image editing. In current generative approaches, however, the slider is simply a rescaling of the method's strength parameter, such as an adapter coefficient, a promp…
- 论文公开站arXiv
One Figure, Every Canvas: Editable Flowchart Relayout via Agentic Pipeline
Pipeline figures in ML papers must be repurposed across many canvases, including paper columns, 16:9 slides, portrait posters, 1:1 social teasers, 9:16 phone previews. Each format imposes a different aspect ratio on the …
- 资讯公开站Digitimes
Winbond, Phison bonuses signal fiercer memory talent war
Memory chip profitability has surged as core operations improve sharply, with some Winbond employees reportedly receiving NT$500,000 to NT$1 million (US$15,733.41–31,466.83) in cash bonuses, helping reverse the com…
- 资讯公开站EE Times
Trusted AI: Why Intelligence Alone Isn’t Enough
- 论文公开站arXiv
Forecasting from Counterfactual Simulator Rollouts: A Sim2Real Evaluation
Deploying a new decision policy creates a cold-start problem for prediction models whose targets depend on the policy's actions: historical observations reflect earlier policies, while real observations under the new pol…
- 资讯公开站rss_arxiv_cs_ai
Build2SPARQL: A Large-Scale Text-to-SPARQL Benchmark Dataset for Building Knowledge Graph Querying
arXiv:2610.00224v1 Announce Type: new Abstract: Building automation systems are increasingly represented as semantic knowledge graphs (KGs) using ontologies such as Brick and ASHRAE 223P, creating a machine-readable sub…
- 资讯公开站rss_arxiv_cs_ai
Gradient-Aligned Pair Selection for Personalized Preference Optimization
arXiv:2610.00061v1 Announce Type: new Abstract: Personalizing large language models (LLMs) requires aligning generation behavior with user-specific preferences rather than aggregate quality. While Direct Preference Opti…
- 资讯公开站Digitimes
OpenAI fires 3 safety researchers for sharing confidential information
OpenAI fired three safety researchers after they shared confidential information with a third-party artificial intelligence safety group, according to The Wall Street Journal, which first reported the news. The case has …
- 论文公开站arXiv
SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation
High-resolution 3D generation increasingly relies on voxel latents and multi-stage pipelines that first predict active structure and then synthesize local geometry. While effective, this design fragments continuous surfa…
- 论文公开站arXiv
Embedding Prediction Helps Image Generation
In diffusion transformers, a class label or a text prompt is embedded once, and the same condition is reused at every denoising step. We ask whether predicted embeddings can serve as this condition instead. Next-Embeddin…
- 论文公开站arXiv
One Basis to Animate Them All: Gaussian Blendshape Distillation for Real-Time Avatars
3D Gaussian avatars support fast rendering, however, their real-time animation is often challenged by the costly neural inference. We address this bottleneck and show that the animation of pretrained avatar models can be…
- 资讯公开站Electrek
E-quipment highlight: 16-tonne Hyundai 160B-X battery-electric forklift
A fresh batch of battery-electric forklift reveals from Hyundai completes the brand’s heavy forklift range, allowing Hyundai dealers to confidently offer a high-voltage electric counterbalance forklift in 4-to-18 tonne c…
- 资讯公开站rss_arxiv_cs_ai
SimTrace: Grounded Multimodal User Trajectories Generation for Online User Modeling
arXiv:2609.38397v1 Announce Type: new Abstract: Virtual clients offer a cost-effective approach to support applications such as A/B testing, recommender system development, and interface evaluation. However, building th…
- 资讯公开站rss_arxiv_cs_ai
Decode-Latency Feedback Prefill: A Model-Free Controller and Its Generalization Limits
arXiv:2609.38386v1 Announce Type: new Abstract: Concurrent autoregressive inference creates a fundamental interference problem: prefilling a newly arrived long prompt can delay tokens for requests that are already decod…
- 论文公开站arXiv
ViTeX-Bench：高保真视频场景文本编辑基准
ViTeX-Bench: Benchmarking High-Fidelity Video Scene Text Editing
视频生成日益逼真可控，但视频编辑仍欠成熟，尤其是需保持原始场景动态的精确局部编辑。视频场景文本编辑需替换店面招牌等场景表面上的文字。
- 资讯公开站rss_arxiv_cs_ai
GeoWind2Plan：面向节能无人机规划的任务时3D城市风场预测
GeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning
在城市低空飞行中，建筑物将环境风重塑为空间变化的三维流场，使无人机（UAV）能耗不仅取决于路径长度，还取决于局部风暴露。然而，当任务必须规划时，建筑物分辨率的风场信息往往不可用。计算流体力学（CFD）可生成高保真城市流场，但每次仿真都绑定于固定的来流边界条件，且可能耗时数小时至数天，这与通常仅持续数分钟到数十分钟的城市无人机任务不兼容。我们提出 GeoWind2Plan，一个用于任务时三维城市风场预测与节能无人机规划的“几何—风场—规划…
- 资讯公开站Digitimes
Anthropic数十亿美元循环融资揭秘：循环资金如何推动AI热潮
Inside Anthropic's multi-billion-dollar loop: how circular financing powers the AI boom
路透社看到的Anthropic机密IPO文件显示，这家AI公司高度依赖少数科技巨头。其业务核心是一个日益循环的金融网络，公司间相互投资与采购形成闭环。
- 论文公开站arXiv
重新思考世界-动作建模的表征
Rethinking Representations for World-Action Modeling
世界-动作模型联合学习机器人策略并预测未来观测，使表征空间成为控制与预测的接口。本文通过受控比较研究该空间的设计，发现重建保真度和预训练均非关键。
- 资讯公开站Energy Storage News
英国7月综述：Engie投运100MW，Fidra与Masdar达成FID，Eku大笔收购，Aura与Starlight获许可
UK July Roundup: Engie commissions 100MW, Fidra and Masdar FIDs, Eku buys big, consents for Aura, Starlight
7月英国大型电池储能活动频繁：Engie投运系统，Masdar与Fidra项目达成财务关闭，Eku Energy收购项目，Aura Power和Starlight Energy获项目许可等。
- 资讯公开站Digitimes
Anthropic 2 万亿美元 IPO 押注附带 80 页警告
Anthropic's US$2 trillion IPO bet comes with 80 pages of warnings
路透社查阅的 Anthropic 首次公开募股（IPO）招股说明书机密细节显示，这家企业正以空前的规模运营，并押注人工智能将比工业革命更深刻地改变全球经济。
- 论文公开站arXiv
如何循环MoE：展平专家，解绑注意力
How to Loop MoE: Flatten the Experts, Untie the Attention
循环Transformer多次复用同一层块：通过额外计算让固定规模模型走得更远，更充分利用参数；而稀疏混合专家（MoE）模型每个token只激活少数专家。循环MoE将两者结合。
- 论文公开站arXiv
FurE：无需动物毛发数据集的高效实例级3D毛发重建
FurE: Efficient Instance-Specific 3D Fur Reconstruction without Animal-Fur Datasets
从多视角图像重建逼真可编辑的动物毛发面临精细细节、自遮挡与混淆等挑战，且不同于人类头发，缺乏动物毛发数据集。
- 资讯公开站rss_arxiv_cs_ai
少想反而模拟更好：直觉式提示提升LLM智能体模拟个体社交媒体反应（含陌生内容）
Thinking Less to Simulate Better: Intuitive Prompting Improves LLM Agents Simulating Individual Social Media Reactions, Including Unfamiliar Content
arXiv:2609.30563v1 公告类型：新 摘要：平台政策越来越多地在人工用户上进行测试，因此智能体的保真度变得重要。然而，有说服力的虚假档案也可能在选举前操纵公众舆论感知。验证一直集中在与人类行为的一致性上，很少关注智能体是否按照其被赋予的档案行事。本研究通过问卷、深度访谈和书面自我介绍对八名塞尔维亚参与者进行了画像，记录他们对六十八篇社交媒体帖子的反应，并让四个语言模型在五种提示条件下预测这些反应，这些条件在档案内容和指令风…
