Results for “model”

The Verge AI Robert Hart SI

Google releases Gemini 4 Carbon with agent ‘swarm’ feature

Google is rolling out its latest AI model, Gemini 4 Carbon, to the general public after releasing it to a select group of "cybersecurity defenders" two weeks ago. Google claims Carbon is its "strongest model" yet, and capable of competing with the very best systems from Anthropic and OpenAI. Google says Gemini 4 Carbon "pushes

Read at The Verge AI

arXiv (research papers) Mehmet Onur Keskin

Personalized Affective Perception via Continual Learning in Human-Agent Negotiation

Self-interested intelligent agents that negotiate with humans should interpret nonverbal expressions and adapt their behavior accordingly. However, generic facial-affect models degrade across individuals and contexts, creating a generalization gap. We introduce an affect-adaptive negotiation strategy that incorporates Continual Learning with Imagination for Facial Expression Recognition (CLIFER) to personalize valence-arousal perception in real time. Differential feedback…

Read at arXiv (research papers)

ZDNET AI Nina Raemont SI

Your FitBit Versa and Sense smartwatches will lose some features this month – what to know

Time’s running out for some Versa and Sense models. After Nov. 28, you can no longer install or update third-party apps or clockfaces on these Google smartwatches.

Read at ZDNET AI

TechCrunch AI Ram Iyer SI

Google broadens its AI hardware portfolio with two new specialized chips

Google's new tensor processing units (TPUs) broaden its chip portfolio to more use cases: confidential model training and more memory capacity.

Read at TechCrunch AI

MarkTechPost Asif Razzaq SI

Zyphra Releases ZONOS3-Mini: Open-Source TTS With 100M Active Parameters

Zyphra released ZONOS3-Mini, an Apache 2.0 text-to-speech model with 800M total and 100M active parameters. It clones voices, steers delivery word by word and runs in real time in the browser

Read at MarkTechPost

arXiv (research papers) Felix Härer Security

Model-Based Authoring of SOC Simulation Exercises for Cybersecurity Education

Cybersecurity training and classroom exercises with a virtual Security Operations Center (SOC) combine alerts, evidence, timeline entries, learner tasks, and assessment items that must remain consistent across an evolving incident. This paper presents a model-based authoring method in which a technical and educational specification is used to construct an incident exercise model and a large language model (LLM) generates selected scenario text based on that model. The…

Read at arXiv (research papers)

The Hacker News Security

Fireworks AI Breach Widens to Expose Customer Billing Data and Incomplete Model Weights

Fireworks AI disclosed on Monday that its investigation into a security incident first reported early last month uncovered further malicious activity, affecting account and billing information, alongside secrets, datasets, and incomplete models for a "limited set of accounts." In a notice published on October 10, 2026, the company said it detected unauthorized use of internal Fireworks

Read at The Hacker News

arXiv (research papers) Ang Yang

Disorder-Induced Logarithmic Spreading and Localization of Quantum Fluctuations

Disorder can localize quantum particles, even in the presence of interactions, but its influence on quantum fluctuations beyond mean--field dynamics remains poorly understood. Here, we investigate the propagation of quantum fluctuations in weakly interacting disordered bosonic systems by comparing the Bose--Hubbard model with the Gross--Pitaevskii equation. We find that, in contrast to ballistic spreading in the disorder--free system, quantum fluctuations exhibit logarithmic…

Read at arXiv (research papers)

arXiv (research papers) Wenjian Hao

Koopman Model-Based Reinforcement Learning

We propose a Koopman model-based reinforcement learning (K-MBRL) algorithm for controlling nonlinear robotic systems with unknown dynamics. The K-MBRL learns a control policy by minimizing an approximate policy objective that combines the analytical immediate cost with an estimate of the future cost evaluated at the one-step predicted state. To this end, we develop a stochastic Koopman operator-based learning method to approximate the unknown nonlinear dynamics in a linear…

Read at arXiv (research papers)

Ars Technica Ryan Whitwam SI

DeepMind reveals Gemini Robotics 2.5 and open safety benchmark "ASIMOV-jailbreak"

The new model is up to twice as successful at some tasks.

Read at Ars Technica

MarkTechPost Asif Razzaq SI

Anthropic Launches Claude Fable 5.2 and Mythos 5.2: Same Model, Different Safeguards

Anthropic released Claude Fable 5.2 and Claude Mythos 5.2, one model with two safeguard levels. Fable 5.2 scores 49.1% on FrontierCode 2.0, costs $8/$40 per million tokens, and ships with a 1M-token context window.

Read at MarkTechPost

Help Net Security Mirko Zorz Security

AI reverse engineering let researchers add a new feature to Slack

Intezer researchers used AI agents to rebuild source code for Slack, Apple’s Finder and a TinyTurla-NG sample, from a backdoor family Cisco Talos attributes to the Turla group. They worked from the shipped programs and vendor documentation. The rebuilt Slack ran well enough that the team added a new feature to it: AI text autocomplete, powered by a small local model. Software vendors that count on compiling their programs to keep source code private are … More →

Read at Help Net Security

arXiv (research papers) Yuhong Wang Security

GARD: Gradient-Guided Adversarial Representation Defense against Backdoors in Multimodal Contrastive Learning Models

Multimodal contrastive learning models such as CLIP are vulnerable to backdoor attacks with diverse trigger structures, ranging from localized patches to globally distributed perturbations, with recent attacks increasingly optimizing these triggers in the representation space. Existing post-hoc defenses based on clean fine-tuning may leave backdoors active, while trigger-inversion methods depend on accurately estimating the attack target and trigger. We observe that…

Read at arXiv (research papers)

arXiv (research papers) Qinfeng Zhu

LEASH: Stay Safe in Agentic Robotics

Foundation models such as GPT and Claude have shown strong capabilities in spatial understanding and embodied intelligence. These advances make agentic robotics a promising path toward artificial general intelligence in robotics. However, we find that using foundation models to control robots poses serious safety risks, and comprehensive safety evaluation is still lacking. We introduce RoboRisk, a comprehensive benchmark for evaluating safety in agentic robotics. Its 92…

Read at arXiv (research papers)

arXiv (research papers) Dimitris Panagopoulos

Model-Free Runtime Measurement of Task-Relevant Information

Autonomous systems increasingly monitor physical kinematics, environmental states and task completion outcomes. While existing work provides methods to plan information seeking actions and interpret acquired observations, current approaches do not monitor whether the live process of information gathering is functioning properly. Consequently, such systems can remain operationally active without resolving outstanding mission requirements. To address this limitation, we…

Read at arXiv (research papers)

SiliconANGLE AI Duncan Riley SI

KORA builds AI models and agents into a new Linux OS bound for Qualcomm Snapdragon PCs

Desktop software company KORA Systems Inc. today unveiled KORA, a Linux-based operating system that builds artificial intelligence models and agents into the system itself. Qualcomm Technologies Inc. has spent close to a year working with the company to bring it to Snapdragon-powered PCs. The launch goes after what KORA sees as a ceiling on AI

Read at SiliconANGLE AI

The Verge Andrew Liszewski Gadgets

Aura’s next color E Ink photo frame is smaller, cheaper, sharper, and more colorful

Aura just launched a smaller version of its color E Ink frame with a handful of improvements and a more affordable price. The new Aura Ink frame features a 10-inch E Ink panel that can display more colors at a higher pixel density. It's also differentiated from the original 13-inch model with a black frame

Read at The Verge

WIRED AI Reece Rogers SI

This New Linux-Based Desktop Operating System Is Built for AI Nerds

Built by Cyanogen’s cofounder, Kora is a next-gen alternative to Windows and macOS built for power users running local AI models.

Read at WIRED AI

TechCrunch Startups Jagmeet Singh Startups

After challenging Google’s Android with Cyanogen, its co-founder takes aim at Windows for AI era

Kora is set to launch as a free-to-use alternative to Windows, targeting users who rely heavily on AI models and agents.

Read at TechCrunch Startups

arXiv (research papers) Tomer Solberg Security

Sable: A Small-Field Sumcheck Argument for Bit-Exact Inference of Large Language Models

We present Sable, an argument system and single-GPU prover for the statement that a committed, integer-quantized transformer produced a given token sequence from a given prompt, bit for bit. Sable fixes an integer function F_O with no floating-point step, in which every rounding, normalization, softmax and sampling decision is determined by its inputs, and proves it with a constraint system whose relations provably never wrap modulo the 31-bit KoalaBear prime. The proof…

Read at arXiv (research papers)

arXiv (research papers) Rezwan Islam Salvi SI

HandsOnAid: Building and Evaluating an Offline First-Aid Assistant for Emergency Response Scenarios

Emergencies can occur where there is no phone signal and no ambulance nearby. A small language model on a phone can be helpful in such situations, but its advice needs to be safe and accurate. We built HandsOnAid, a 4-bit Gemma 2B model fine-tuned with QLoRA on FirstAidQA dataset, a corpus of 5,550 first-aid question-answer pairs. We compared seven configurations, including the base, 4-bit and 8-bit fine-tuned models, and versions with a length floor, safety gate, or…

Read at arXiv (research papers)

arXiv (research papers) Zhikun Tao SI

Can KV Cache Be Reused Across Model Updates? A DeepSeek-V4-Flash Case Study

Reusing key--value (KV) caches across language model updates could avoid repeated computation of shared context, but changes in model weights raise a fundamental question about what these caches preserve. We investigate direct cache reuse across four model updates, with a detailed study of DeepSeek-V4-Flash to Flash-0731. Our central finding is that caches can retain useful task information without faithfully reproducing the updated model's behavior. Retrieval accuracy…

Read at arXiv (research papers)

arXiv (research papers) Xuheng Zhang

LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning

World Action Models (WAMs) have recently emerged as a promising paradigm for robotic manipulation, endowing policies with strong capabilities for modeling temporal and physical dynamics. Existing WAMs, encompassing both video-first action prediction and joint video-action modeling, directly transfer global physical representations from world models to action generation. However, global scene representations may underemphasize local robot-object interactions, creating a…

Read at arXiv (research papers)

South China Morning Post Brian Wong Geopolitics

Abby Choi’s ex-brother-in-law tells court he was ‘a rather special existence’

The former brother-in-law of slain Hong Kong socialite Abby Choi Tin-fung has insisted in court that the model was “not in a position” to blame him for her strained relationship with his family before her alleged murder, citing their friendship and his contribution to her children’s upbringing. Anthony Kwong Kong-kit on Tuesday portrayed himself as “a rather special existence” between the two families, saying that he had been a “very good friend” of the model before he…

Read at South China Morning Post

arXiv (research papers) Jiahao Wang SI

Program-Aided Patchable Skill Evolution

Agent skills have emerged as an effective means of enabling LLM agents to solve complex, interactive tasks by encoding reusable knowledge without modifying model parameters. Recent approaches further improve skill adaptation by distilling reusable experience from agent execution trajectories and evolving skills based on downstream task feedback. However, existing methods primarily represent skills as natural language documents, intertwining procedural logic with descriptive…

Read at arXiv (research papers)

arXiv (research papers) Chi Liu SI

Margin-Guided Token Credit for Reinforcement Learning with Verifiable Rewards

In reinforcement learning with verifiable rewards, a language model is rewarded for a correct final answer, and GRPO passes that reward to every token of the solution in equal measure. We instead place the credit where the model hesitated, in a method we call MCPO (Margin-Credit Policy Optimization). We measure hesitation by the top-2 margin $M=p_1-p_2$ of the policy's next-token distribution, map it to a score with two fixed thresholds, and spread each solution's advantage…

Read at arXiv (research papers)

arXiv (research papers) Lirui Luo SI

Coding Agent Memory Post-Training II: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via On-Policy Distillation

Language-model agents increasingly tackle long-horizon tasks whose interaction histories exceed the model's active context. They must therefore decide what to write down and what to read back, and when to do either. Recent work has begun to use on-policy distillation (OPD) to make memory control part of the policy, relying on memory tools and compaction protocols that lie outside either model's pre-training. However, this weakens the distillation on both sides: the teacher…

Read at arXiv (research papers)

arXiv (research papers) Edward Jiang

Normalizing Flows for Classical Simulation of Quantum Circuits in Phase Space

We present a novel framework to learn a representation of quantum circuits based on the phase space formulation of many-body systems. Given that $N$-qubit states can be represented as probability distributions on a smooth $2N$-dimensional manifold, we show how unitary gates can be parameterized as normalizing flows between these distributions, opening a new avenue for quantum circuit simulation using generative models. We evaluate our method on three common families of…

Read at arXiv (research papers)

arXiv (research papers) Xingqi Wang SI

Specify Privacy Yourself: Assessing Inference-Time Personalized Privacy Preservation Ability of Large Vision-Language Models

Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities but raise significant privacy concerns due to their abilities to infer sensitive personal information from images with high precision. While current LVLMs are relatively well aligned to protect universal privacy, e.g., credit card data, we argue that privacy is inherently personalized and context-dependent. This work pivots towards a novel task: can LVLMs achieve Inference-Time Personalized…

Read at arXiv (research papers)

Digital Trends Vikhyaat Vivek

The iPhone Air 2 could finally get the camera upgrade it desperately needed

Alleged dummy units of Apple's upcoming iPhone Air 2 and iPhone 18 offer an early glimpse at their designs, including a second rear camera for the Air and a familiar camera arrangement for the standard model.

Read at Digital Trends

Older stories