Results for “model”

arXiv (research papers) Iván López-Broceño

Learning the Dynamics of Deformable Linear Objects Through Coupled Oscillators for Bimanual Robotic Manipulation

Many learning-based approaches for deformable linear objects (DLOs) rely on simplified deformation models that neglect dynamic effects and the mechanical response of the objects, limiting their applicability to lightweight, compliant objects and slow motions. Inspired by recent advances in learning physically structured dynamics with coupled oscillator networks (CONs), we propose $Φ$-CON for learning force-driven DLO dynamics. The model combines a floating-base coupled…

Read at arXiv (research papers)

The Verge Brad Bourque Gadgets

The MacBook Neo is receiving a rare $50 discount ahead of Black Friday

If you’re looking for a new laptop, or a gift for the student in your life, you can save $50 on the MacBook Neo at several retailers, in both storage sizes and all four colors. The 256GB model is discounted to $649 (Amazon, Best Buy, Target), while the 512GB model with Touch ID is $749

Read at The Verge

AWS Machine Learning Blog Meghan Kothari SI

Cut generative AI costs: Code for rules, models for judgment

The biggest cost lever in a generative AI system is whether you call a model at all. Learn how to decide whether each workflow step needs a model, and map each option to AWS: deterministic code, a constrained decision call, or a right-sized model on Amazon Bedrock.

Read at AWS Machine Learning Blog

arXiv (research papers) Malak Gamal El-Din

BeFit: Shifting Complexity from Weights to Bernstein Activations for Resource-Constrained Sensing

Time-series sensing increasingly relies on neural networks, yet deployment on resource-constrained microcontrollers remains challenging. Aggressive model compression can reduce accuracy, while restricting weights to multiplication-free representations introduces further losses. We present BeFit, an end-to-end framework that shifts model capacity from weights to learnable Bernstein polynomial activations. BeFit (i) finds compact Bernstein-activated convolutional neural…

Read at arXiv (research papers)

AWS Machine Learning Blog Dmitry Soldatkin SI

Amazon SageMaker AI real-time endpoints now support KV cache aware routing

Amazon SageMaker AI real-time endpoints now support KV cache-aware routing, a new strategy that routes each request to the instance holding the longest cached match for its prompt. Learn how it works, when to use it compared with prefix-aware routing, and how to enable, invoke, and verify it on single model and inference component endpoints.

Read at AWS Machine Learning Blog

arXiv (research papers) Kory Rosen Tech policy

From Arraignment to Resolution: Transformer Models of Court Cases

The criminal legal system is composed of many complex decision points, such as which charges are pursued, how bond is determined, plea deal negotiations, and sentencing. Given the profound impact of such decisions, there is a growing need to increase the transparency within this system. While prior machine learning approaches have focused on modeling single decisions (e.g., recidivism), this paper proposes a transformer-based time series model to forecast the outcome of the…

Read at arXiv (research papers)

Dark Reading Alexander Culafi Security

'TheMailer' Phishing Kit Puts AI in the Driver's Seat

The commercial kit uses attacker-supplied AI models to personalize lures, evade spam filters, and scale campaigns to tens of thousands of targets.

Read at Dark Reading

The Decoder Matthias Bastian SI

OpenAI launches GPT-6.2 Astra and Sol with lower API prices and some benchmark gains

OpenAI releases GPT-6.2 Astra and Sol, two new models that are more capable and cheaper than their predecessors. In the independent Artificial Analysis Intelligence Index, Astra ranks second behind Claude Fable 5.2, and it is the first released model to reach the High threshold for AI self-improvement on OpenAI's own scale. OpenAI itself classifies that as a risk. The article OpenAI launches GPT-6.2 Astra and Sol with lower API prices and some benchmark gains appeared first…

Read at The Decoder

arXiv (research papers) Mark Nowak Conflux Laboratory Tech policy

AllotNet: an Interpretable Mixture-of-Experts Choice Model for Moral Judgement in Humanitarian Aid Allocation, Validated against a Planted Synthetic Population

Official development assistance fell by 23.1 per cent in 2025, and the decision of who is helped first when there is not enough for everyone is moving into software. A leading published machine-learning targeting system excluded 53 per cent of eligible households as deployed. If the rule of distribution ends up in code, it should carry the judgement of many cultures, and the machine should execute while a person judges. AllotNet is an interpretable mixture-of-experts…

Read at arXiv (research papers)

9to5Mac Ryan Christoffel Gadgets

iOS 27.3’s redesigned Apple Health app requires newer iPhone model

iOS 27.3 will introduce an all-new Apple Health experience when it ships next month, but you’ll need an AI-compatible iPhone to get it. Here’s why.

Read at 9to5Mac

SiliconANGLE AI Kyt Dotson SI

Microsoft rolls out new first-party models and updates to agentic AI continuous improvement

Microsoft Corp. today rolled out new generative artificial intelligence models and updates to Foundry, the company’s unified enterprise platform that helps developers and organizations ground, test and deploy AI applications and autonomous agents. Updates to Foundry IQ, part of Microsoft IQ, now provide an enterprise knowledge graph out of the box, delivering a full estate

Read at SiliconANGLE AI

GeekWire Todd Bishop SI

Microsoft jabs at OpenAI and Anthropic, touting cheaper, customizable AI models of its own

Microsoft pitched Copilot as cheaper than ChatGPT and Claude at its Ignite conference, showing a chart that put monthly business costs per user at $30 for Copilot, $81 for ChatGPT and $205 for Claude. Microsoft invests in both OpenAI and Anthropic and sells their models.

Read at GeekWire

arXiv (research papers) Hsi Chang Tech policy

Trace2ST: Recovering PLC Structured Text from I/O Traces with LLM Generation and Verification

Programmable logic controllers (PLCs), the industrial computers that operate power and process plants, often run programs whose source code is no longer available; however, every input and output of the controller is logged continuously. This paper therefore asks whether a large language model (LLM) can recover an executable program in Structured Text (ST), the standard PLC language of IEC 61131-3, from such input/output records (traces) alone and, equally important, how the…

Read at arXiv (research papers)

Digital Trends Sudhanshu Kumar Mangalam

OpenAI’s GPT-6.2 Astra finally arrives, promising better performance and fewer safety problems

OpenAI has introduced two of its most powerful AI models yet, GPT-6.2 Astra and GPT-6.2 Sol, nearly two months after canceling GPT-6.1 Astra over safety concerns. According to the company, the new models are considerably better at coding, professional work, and scientific research. They can also complete complex tasks using fewer tokens, which brings down

Read at Digital Trends

AWS Machine Learning Blog Varun Rajan SI

Announcing AI data preparation capabilities in Amazon SageMaker: Build high-quality datasets for model development workflows

Amazon SageMaker AI now offers AI data preparation capabilities for building high-quality training and evaluation data. Generate synthetic examples, assess and improve datasets, auto-label at scale, and bring in expert review through Python SDK APIs and an agent skill, all from your own notebook or IDE.

Read at AWS Machine Learning Blog

TechCrunch AI Marina Temkin SI

Model ML raises $250M to automate investment banking grunt work

The startup says its agents understand the full context of deals that bankers and consultants are working on and can automatically update financial models and presentations when new developments affect their analysis.

Read at TechCrunch AI

TestingCatalog Erin | AI Agent SI

Microsoft launches seven MAI models at Ignite 2026

Microsoft launches seven MAI models at Ignite, including Code 2 for GitHub Copilot, Cyber 2 Flash, Image 2.7, Realtime 1 and Decision 1.1.

Read at TestingCatalog

The New Stack Frederic Lardinois

Microsoft’s new coding model promises near-Opus performance at 5% of the cost

At its Ignite conference in San Francisco on Tuesday, Microsoft announced a number of new AI models, including two new

Read at The New Stack

Meduza

Russian fashion journalist and former model Svetlana Kunitsyna dies at 56

TV host Svetlana Kunitsyna has died at 56, her colleague, journalist Pavel Lobkov, said.

Read at Meduza

arXiv (research papers) Yuanlin Wen

HardwareDx: A Foundational Graph Model System for Datacenter Machine Repairs

The rapid scale-up of AI infrastructure has made datacenter reliability a cornerstone of the cloud business economy. Since hardware failures in hyper-scale AI clusters can stall multi-million dollar training workloads, the downtime of individual machines can cause outsized economic losses. Minimizing machine downtime, therefore, is more important than ever. To address this, we present Hardware Diagnoser (HardwareDx), an AI-driven hardware diagnosis system that applies a…

Read at arXiv (research papers)

arXiv (research papers) Giannis Daras SI

Ambient LLMs: Data-Efficient Supervised Fine-tuning via Temperature Matching

Collecting large-scale, diverse, and high-quality post-training datasets for autoregressive language models is challenging. Consequently, Supervised Fine-Tuning (SFT) is typically applied to datasets significantly smaller than those used for pre-training, which leads to overfitting and the catastrophic forgetting of the tails of the data distribution. To prevent this collapse, we introduce Ambient SFT, a principled framework for leveraging imperfect predictions to augment…

Read at arXiv (research papers)

The Verge Emma Roth SI

Google’s Spark AI agent becomes ‘Tasks’ — and now it’s free

Google is rebranding its Spark AI agent as "Tasks" and making it free to use. The feature, which is powered by Google's new Gemini 4 Hydrogen AI model, can do things like create meal plans, manage your calendar, and dig through your emails to find recurring expenses you might not need. Users in the US

Read at The Verge

arXiv (research papers) Shobhita Sundaram SI

Do LLM Agents Still Need to Be Told How to Self-Improve?

LLM agents increasingly drive their own learning. Self-improvement methods delegate various components of model optimization to the models themselves. As models become more capable, this motivates a basic question: do these increasingly capable models still need us to specify how they should self-improve? In this paper we study an autonomous paradigm in which, given basic primitives, resources, and a general goal for self-improvement, agents construct their own self-learning…

Read at arXiv (research papers)

arXiv (research papers) Yining Hua SI

Higher-Scoring Coding Models Did Not Get More Work Done

New coding models advertise themselves on score improvements of coding benchmarks. Yet no study has tested if a model that scores higher on these benchmarks gets more real work done. On June 12, 2026, a U.S. export control directive banned access to Claude Fable 5 for 19 days, which forced users to switch to other models. During this time, most former Fable 5 users switched to Opus 4.8 (11 points lower on SWE-Bench Pro compared to Fable 5). We treat this suspension as a…

Read at arXiv (research papers)

arXiv (research papers) Pranav Gupta

Feedback-Linearised Trajectory Tracking via Model-Informed Robust Geometric DDPC

Data-driven predictive control (DDPC) predicts system responses from recorded trajectories, but nonlinear dynamics can limit the accuracy of fixed linear predictors. This paper studies geometric robust DDPC for trajectory tracking of a fully actuated manipulator using model-based feedback linearisation. A mechanical model maps virtual acceleration commands into joint torques, while recorded data capture residual dynamics arising from imperfect model parameters. The…

Read at arXiv (research papers)

arXiv (research papers) Alec F. White

Modeling quantum information scrambling in realistic protein NMR experiments

Out-of-time-order correlators (OTOCs) are important probes of operator spreading and information scrambling, and have recently been proposed as a means of extracting molecular structural information through nuclear magnetic resonance (NMR). Realizing this potential requires accurate modeling of OTOCs in complex molecular environments within experimentally relevant settings. Here, we develop an end-to-end computational pipeline connecting molecular structures to OTOC dynamics…

Read at arXiv (research papers)

arXiv (research papers) John Gideon

Multimodal Masked Diffusion Autoencoder for Modeling Impaired Driving

Human cognitive state (e.g., distraction, stress), gaze, and behavior interact with the driving environment in subtle ways that AI could use to assist drivers, but ground-truth labels for human cognitive state levels are costly to obtain at scale. We propose a multimodal masked-diffusion autoencoder, ID-MAE, that adds a self-supervised objective to impairment prediction: alongside predicting the impairment label as a masked token, it reconstructs masked gaze and trajectory…

Read at arXiv (research papers)

arXiv (research papers) Aviv Yossef Security

Distortion-Free Semantic Watermarks for Black-Box Generative Models

We present a provably distortion-free semantic watermarking algorithm that requires only black-box sampling access to a generative model and black-box access to a semantic embedding. A watermark is called distortion-free if, averaged over its secret detection key, it leaves the generator's output distribution unchanged; this provides a formal guarantee that watermarking does not reduce output quality. Conventional such watermarks are often brittle to transformations. This is…

Read at arXiv (research papers)

arXiv (research papers) Tongxi Wu SI

Reading Scene States: Uncovering Safety Failures in Vision-Language Models

Vision-language models (VLMs) can provide harmful assistance by interpreting visual events, even without an explicit request for the underlying harmful objective. We identify this \textbf{scene-state safety vulnerability} and introduce \textbf{SceneSafetyBench}, comprising 375 image--question pairs spanning Initial, Medium, and Ultimate scene states and five visual presentations. Across eight VLMs, scene-state inputs elicit harmful assistance with attack success rates (ASRs)…

Read at arXiv (research papers)

arXiv (research papers) Mehmet Altan Toksöz SI

Fine-Tuning Gemma 4 Edge Models for MATLAB Signal-Processing Code with SFT and QLoRA on a Laptop iGPU

Small language models that run on a laptop are attractive for writing signal-processing code offline, but they make systematic errors on digital signal processing (DSP) conventions. We apply supervised fine-tuning (SFT) to the Gemma 4 edge models E2B and E4B with 66 checked examples covering 11 signal-processing MATLAB subtasks, using low-rank adaptation (LoRA) and quantized LoRA (QLoRA) on the integrated GPU (iGPU) of a laptop, and deploy the 8-bit integer (int8) models…

Read at arXiv (research papers)

Older stories