今日

category.学术arXiv cs.CV (计算机视觉)

Hybrid Cross-Modal Attention Network for Early Breast Cancer Detection in Low-Resource Clinical Settings

arXiv:2610.07243v1 Announce Type: new Abstract: Breast cancer is the leading cause of cancer-related mortality among women in Sub-Saharan Africa, where delayed diagnosis results from limited radiology expertise and fragmented clinical data systems. Although deep learning models have demonstrated strong performance in mammographic analysis, most rely solely on imaging data and are trained on Wester

10月7日 04:00

category.学术arXiv cs.LG (机器学习)

Neutrosophic Ensemble Classification for Uncertainty-Aware Bearing Fault Detection: Evidence from Laboratory and Variable-Speed Industrial Benchmarks

arXiv:2610.06880v1 Announce Type: new Abstract: Machine learning classifiers for bearing fault detection produce scalar confidence scores that conflate confident errors with genuinely ambiguous predictions, and the conventional truth/falsity pair (F = 1 - T) is algebraically redundant by construction. We operationalize a refined neutrosophic decomposition of a Random Forest + XGBoost + Logistic Re

10月7日 04:00

category.学术arXiv cs.CL (自然语言处理)

Zero-Shot Visualization: Exploring Text Corpora with User-Prompted Axes

arXiv:2610.06889v1 Announce Type: new Abstract: We study the application of large language models (LLMs) to the visual exploration of textual corpora. We introduce zero-shot visualization (ZSV), a task in which users specify concepts in natural language and documents are mapped onto the corresponding concept axes for visualization. Building a ZSV system of practical value is non-trivial, as it req

10月7日 04:00

category.学术arXiv cs.LG (机器学习)

Comparative review of hybrid forecasting models for short-term prediction of building thermal load

arXiv:2610.06881v1 Announce Type: new Abstract: In this paper, a comparative review of different hybrid models for short-term forecasting of building thermal demand is carried out. Particularly, the assessment tackles the comparison of data-driven models enhanced with other state-of-the-art techniques. At the first step, the existing techniques reported in the literature are analysed. It is conclu

10月7日 04:00

category.学术arXiv cs.LG (机器学习)

TEMPEST: Temporal Embeddings for Scalable Driver Identification via Angular Margin Learning

arXiv:2610.06855v1 Announce Type: new Abstract: Scalable driver identification requires embedding models that maintain discriminative performance as fleet size grows, yet existing triplet-loss formulations degrade rapidly with driver pool size and overfit to session-specific patterns under rigorous temporal evaluation. We introduce TEMPEST, a Temporal Convolutional Network embedding model trained

10月7日 04:00

category.学术arXiv cs.LG (机器学习)

Few-Shot Bioactivity Prediction with Meta-Learning under Assay Heterogeneity

arXiv:2610.07079v1 Announce Type: new Abstract: Accurate bioactivity prediction is a central challenge in early-stage drug discovery, as individual assays often contain too few measurements to train reliable models independently. Meta-learning offers a principled approach to this few-shot setting, but assay heterogeneity may limit its effectiveness. Here, we test this hypothesis and show that meta

10月7日 04:00

category.学术arXiv cs.LG (机器学习)

Learning When to Refine: Long-Horizon Reinforcement Learning for Budgeted Neural-Operator PDE Solvers

arXiv:2610.06883v1 Announce Type: new Abstract: Neural operators provide fast surrogates for time-dependent PDEs, but autoregressive deployment creates a refinement-allocation problem: prediction errors vary over space and time, while only a finite number of local corrections can be committed along a trajectory. We formulate this as budgeted adaptive neural-operator solving. A global Fourier neura

10月7日 04:00

category.学术arXiv cs.LG (机器学习)

What Must Replay Preserve? Separating Correctable Bias from Class Correspondence

arXiv:2610.07077v1 Announce Type: new Abstract: Class-incremental learning must recognize all classes seen so far without task labels. Logit replay methods such as DER and DER++ mitigate forgetting by matching the model's past predictions on stored examples. Deleting this matching reveals its benefit, but the resulting accuracy cost cannot show whether the stored scores themselves are needed, or w

10月7日 04:00

category.学术arXiv cs.LG (机器学习)

When Does External Guidance Help LLM Reasoning? A Bias-Variance Theory of Guidance-Augmented GRPO

arXiv:2610.06861v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has become the dominant paradigm for eliciting multi-step reasoning in large language models, and a recent wave of methods (LUFFY, ExPO, PAPO, TAPO) further augments RL with \emph{external guidance} - expert traces, self-explanations, or retrieved thought patterns. Although each method reports emp

10月7日 04:00

category.学术arXiv cs.CL (自然语言处理)

Not What a Child Expressed: Auditing the Sign-to-Text Safety Interface in Child-Facing AI

arXiv:2610.07519v1 Announce Type: new Abstract: Automatic sign language translation (SLT) has entered consumer products, turning American Sign Language into English text for dictation, messaging, and queries put to a conversational assistant. Child-facing AI and platform trust-and-safety tooling decide on text, using filters on minor accounts and grooming classifiers that score chat messages. A si

10月7日 04:00

category.学术arXiv cs.CL (自然语言处理)

Kurate: Scalable Scientific Quality Analysis

arXiv:2610.07306v1 Announce Type: new Abstract: Scientific search systems can find papers that are relevant to a question, but they generally do not assess the quality of the evidence that those papers provide. We present Kurate, a system that uses large language models (LLMs) to assess the quality of published studies. Kurate uses both the paper and its related documents (e.g., the study's trial

10月7日 04:00

category.学术arXiv cs.CL (自然语言处理)

Quality-Aware Self-Correcting Speech Translation on an Edge Device

arXiv:2610.07545v1 Announce Type: new Abstract: We present a fully offline speech-to-speech translation pipeline that runs on a Jetson Nano (4 GB) and corrects its own weak translations without retraining. A Whisper-tiny ASR feeds an Opus-MT translator; multilingual BERT cosine similarity acts as a Quality Estimation (QE) gate, triggering a secondary-pass correction when confidence falls below a p

10月7日 04:00

category.学术arXiv cs.CL (自然语言处理)

Recurrent Looped Transformer

arXiv:2610.07591v1 Announce Type: new Abstract: State tracking requires an update at every input, but the depth a Transformer applies to each token is fixed regardless of sequence length. We introduce the Recurrent Looped Transformer (RLT), which splits its layers between a parallel causal encoder and a recurrent decoder. At each token, the decoder merges the encoder output with the previous token

10月7日 04:00

category.学术arXiv cs.CL (自然语言处理)

Large Language Model Orchestration under Heterogeneous Preferences via Explicit Persona Inference

arXiv:2610.07587v1 Announce Type: new Abstract: LLM orchestration investigates how an orchestrator coordinates a group of autonomous agents to achieve common goals or maximize collective welfare. The agents are typically heterogeneous, each holding a private preference that it pursues but does not reveal. Inferring such hidden preferences from behavior has been a subject of long-standing research

10月7日 04:00

category.科技媒体TechMeme

Japanese chipmaker Rapidus is partnering with 17 companies, including US-based Synopsys, to help customers design chips; Rapidus has $15B+ in state funding (Reuters)

Reuters : Japanese chipmaker Rapidus is partnering with 17 companies, including US-based Synopsys, to help customers design chips; Rapidus has $15B+ in state funding   —  Japan's Rapidus, with $15 billion in state backing, is tying up with chip design firms as it seeks to answer a major question hanging …

10月7日 03:35

category.科技媒体TechMeme

A September 24 cyberattack on Arizona's court system stole PII for 1.3M people, including 30K orders of protection and 150K reports from a foster care board (Associated Press)

Associated Press : A September 24 cyberattack on Arizona's court system stole PII for 1.3M people, including 30K orders of protection and 150K reports from a foster care board   —  A cyberattack on Arizona's court system stole personal information for more than a million people and is believed to have started …

10月7日 03:10

category.科技媒体少数派

罗马：永恒之城，永恒于世

Matrix首页推荐Matrix是少数派的写作社区，我们主张分享真实的产品体验，有实用价值的经验与思考。我们会不定期挑选Matrix最优质的文章，展示来自用户的最真实的体验和观点。文章代表作者个人观点 ...<a href="https://sspai.com/post/114845" target="_blank">查看全文</a>

10月7日 03:00

category.科技媒体钛媒体

好产品也会丢单，Palantir 旧访谈翻红：企业级 AI 订单，拼的不只是模型

人们对一件事的认知，有时比客观情况更能左右决策。

10月7日 02:54

category.科技媒体TechMeme

A US federal court sentences Michael Smith, who pleaded guilty to using bots to stream AI-generated songs, making $8M+ in royalties, to 18 months in prison (Bill Donahue/Billboard)

Bill Donahue / Billboard : A US federal court sentences Michael Smith, who pleaded guilty to using bots to stream AI-generated songs, making $8M+ in royalties, to 18 months in prison   —  Michael Smith admitted earlier this year to using bots to stream fake songs billions of times in a first-ever criminal case over streaming fraud.

10月7日 02:20

category.科技媒体TechMeme

Boston Dynamics appoints former Amazon executive Rohit Prasad, who spent 12 years helping build and expand Alexa, as its CEO, after Robert Playter stepped down (Samantha Kelly/Bloomberg)

Samantha Kelly / Bloomberg : Boston Dynamics appoints former Amazon executive Rohit Prasad, who spent 12 years helping build and expand Alexa, as its CEO, after Robert Playter stepped down   —  Boston Dynamics appointed former Amazon.com Inc. executive Rohit Prasad as chief executive officer to lead the Hyundai Motor Co …

10月7日 01:30

category.科技媒体量子位

OpenAI一夜甩出722篇数学论文！黎曼霍奇BSD全上阵，数学家：读不过来

三位菲尔兹奖得主：不代表认可

10月7日 01:05

category.科技媒体TechMeme

GLM-5.3's open release has yet to produce major attacks despite Anthropic's warnings about its Mythos-level cyber risk, undercutting calls to ban open models (Nathan Lambert/Interconnects AI

Nathan Lambert / Interconnects AI : GLM-5.3's open release has yet to produce major attacks despite Anthropic's warnings about its Mythos-level cyber risk, undercutting calls to ban open models   —  Open-weights, ideology, and acknowledging trade-offs. … Audio playback is not supported on your browser.  Please upgrade.

10月7日 01:05

category.科技媒体钛媒体

默沙东重注背后：G12D凭什么成为下一个爆款靶点？

默沙东为何愿意为了一款早期药物一掷千金呢？

10月7日 00:56

category.开发者Simon Willison's Blog

OpenAI “rogue” agent activities found on Wikimedia projects

<a href="https://wikimediafoundation.org/news/2026/10/05/openai-rogue-agent-activities-found-on-wikimedia-projects/">OpenAI “rogue” agent activities found on Wikimedia projects</a> Given how tempting a target wikis are for rogue agent swarms, it's not a huge surprise that Wikipedia found evidence of that activity once they went looking: <blockquote> The Wikimedia Fou

10月7日 00:16

category.科技媒体TechMeme

New York-based Melius, which provides AI tools for generating ad campaigns, images, and videos, raised $25M, including a $20M Series A and a $5M seed (Marina Temkin/TechCrunch)

Marina Temkin / TechCrunch : New York-based Melius, which provides AI tools for generating ad campaigns, images, and videos, raised $25M, including a $20M Series A and a $5M seed   —  Melius, an AI platform for generating ad campaigns, images, and videos, announced on Tuesday that it raised a total of $25 million in funding …

10月7日 00:15