The Cold-Start Safety Gap in LLM Agents
arXiv:2606.07867v2 Announce Type: replace Abstract: Are tool-calling LLM agents equally safe throughout a conversation? We discover they are not: agents are most vulnerable at the very start of a sess
Knowledge catalogue
arXiv:2606.07867v2 Announce Type: replace Abstract: Are tool-calling LLM agents equally safe throughout a conversation? We discover they are not: agents are most vulnerable at the very start of a sess
arXiv:2608.26895v1 Announce Type: cross Abstract: Agentic recommender systems ground each decision of a large language model (LLM) in a persistent memory of the user, and in existing agents that memor
arXiv:2608.25489v1 Announce Type: cross Abstract: Graph retrieval-augmented generation places retrieved subgraphs into the model's context window at query time, paying a recurring token cost and expos
arXiv:2608.25948v1 Announce Type: new Abstract: Medical image foundation models can predict clinical phenotypes from computed tomography (CT), but strong performance leaves open whether they read dise
arXiv:2608.25480v1 Announce Type: new Abstract: Dataset distillation compresses a large training set into a compact synthetic set while preserving its downstream utility. However, existing methods pri
Need a low-cost, high-performance way to run long-lived, stateful workloads such as AI agents? Today, we introduced Cloud Run instances, which let you do just that. Consider AI agents such as OpenClaw
arXiv:2608.25305v1 Announce Type: new Abstract: Training-free zero-shot composed image retrieval finds a target image in a gallery from a reference image and a text edit without learning from task-spe
Maxwell Zeff / Wired: OpenAI is testing a “Persistent mode” in Codex, designed to let AI agents “continue working until put to sleep” and proactively generate follow-up tasks — Code reviewed by WIRED
arXiv:2608.24825v1 Announce Type: new Abstract: The rapid expansion of large-scale assessments and the growing adoption of automatic item generation have intensified concerns about incidental content
arXiv:2608.23569v1 Announce Type: new Abstract: State-of-the-art Natural Language to SQL (NL2SQL) models report execution accuracy exceeding 89 percent on established benchmarks such as Spider and BIR
GLM-5.3-Flash scores 57 on the Artificial Analysis Intelligence Index. At 0.09 Cost per Task, it sits comfortably on the Intelligence vs. Cost per Task Pareto frontier @Zai_org has released GLM-5.3-Fl
arXiv:2608.24541v1 Announce Type: new Abstract: Surgical instrument segmentation (SIS) is fundamental for computer-assisted surgery, where reliable instrument masks enable precise scene understanding
arXiv:2608.23745v1 Announce Type: cross Abstract: Accurate brain tumor segmentation from magnetic resonance imaging (MRI) is essential for diagnosis, treatment planning, surgical guidance, and disease
arXiv:2608.23906v1 Announce Type: new Abstract: Artificial Intelligence (AI) is increasingly integrated into complex sociotechnical systems, including Critical National Infrastructure (CNI), where har
arXiv:2509.03754v2 Announce Type: replace-cross Abstract: Responding to rising global food security needs, precision agriculture and deep learning-based plant disease diagnosis have become crucial. Ye
arXiv:2608.22679v1 Announce Type: new Abstract: Semantic segmentation has rapidly advanced with deep learning; however, challenges remain in effectively capturing local and global contexts as well as
arXiv:2608.21946v1 Announce Type: cross Abstract: Reinforcement learning with outcome-based objectives such as GRPO enables LLM-based agents to solve complex, long-horizon tasks, yet the reusable expl
arXiv:2608.22559v1 Announce Type: new Abstract: Rubrics aim to make language-model evaluation transparent by decomposing response quality into interpretable criteria. However, natural-language rubrics
arXiv:2606.17642v3 Announce Type: replace Abstract: Financial multimodal reasoning requires agents to coordinate numerical computation, retrieval, visual interpretation, and temporal grounding across
arXiv:2608.21786v1 Announce Type: new Abstract: General image fusion seeks to integrate complementary information from multiple source images, yet real-world applications often require a single system
arXiv:2608.22500v1 Announce Type: new Abstract: At the heart of composed visual data retrieval is the fusion of a reference visual input and a textual modification into a single query. While current s
arXiv:2608.22370v1 Announce Type: new Abstract: Task-specific LoRA adapters offer a modular way to specialize large language and vision-language models. However, existing adapter composition methods a
arXiv:2608.23055v1 Announce Type: new Abstract: Navigating large, photorealistic 3D apartments from raw pixels is widely considered infeasible for plain reinforcement learning. We build an agent that
arXiv:2608.21867v1 Announce Type: new Abstract: LLM agents are moving from single-prompt use to long task streams in which reusable memory becomes a core capability for terminal, software-engineering,
arXiv:2608.23565v1 Announce Type: new Abstract: An interactive world model must follow the user's actions, remember the places it has shown, and stream in real time. The tension is structural: control
I’ve been experimenting with a custom inference/runtime research project called CRANE V2, mostly because I wanted to answer a stupid question: How far can you push absurdly large MoE models on an ordi
arXiv:2608.22538v1 Announce Type: new Abstract: Policy-governed agents must interpret case evidence while following an authorized procedure. We present extsc{Stage}, an executable-graph framework that
arXiv:2510.09930v2 Announce Type: replace-cross Abstract: Real-world systems, ranging from industrial manufacturing to wearable healthcare, generate multivariate time series with hierarchical states r
arXiv:2608.20825v1 Announce Type: new Abstract: Artificial intelligence systems increasingly make consequential judgments - which patient is deteriorating, which building is safe to enter, whether an
arXiv:2608.21098v1 Announce Type: new Abstract: Fusing prior knowledge with data-driven learning is attractive where data is scarce, yet no controlled account says when it helps, is redundant, or harm
arXiv:2608.19817v1 Announce Type: cross Abstract: Conventional convolutional kernels are typically defined on fixed discrete grids, limiting their ability to accommodate heterogeneous local structures
arXiv:2607.21596v2 Announce Type: replace Abstract: Large language model agents can adapt to complex tasks by constructing workflows at inference time, but procedures discovered in one episode are usu
arXiv:2608.19430v1 Announce Type: cross Abstract: Vulnerability prioritization is inherently preference dependent, since the same CVE can receive different remediation priority under different operati
arXiv:2608.19625v1 Announce Type: new Abstract: Scientific data are increasingly used by AI agents, yet existing dataset representations provide limited support for autonomous discovery, interpretatio
It’s never been easier to start an AI-powered startup on Google Cloud. You grab an API key from Google AI Studio at breakfast, paste it into Antigravity, and by lunch you’ll have a nascent prototype o
arXiv:2608.18654v1 Announce Type: new Abstract: Medical image captioning requires translating heterogeneous visual evidence into concise clinical descriptions, where errors in findings, assertion stat
We are thrilled to announce that Google has been recognized as a Leader for the third year in a row in the 2026 Gartner® Magic Quadrant™ for Cloud-Native Application Platforms (CNAP). We believe this
I build LoRA Dataset Studio — free, open source, self-hosted, no account and no telemetry. It is not a competitor to ai-toolkit: it orchestrates it. ai-toolkit is the trainer; this is everything befor
arXiv:2608.18171v1 Announce Type: new Abstract: Looped language models have shown promising results on reasoning benchmarks, yet their potential for agentic tool use remains largely unexplored. We stu
arXiv:2608.18952v1 Announce Type: cross Abstract: Large language models can improve recommendation quality by reasoning explicitly over user history and candidate items - for example, extracting a use
arXiv:2608.18585v1 Announce Type: new Abstract: Vision-based industrial anomaly detectors are calibrated on one distribution but may be deployed on another that differs in illumination, fixture placem
arXiv:2608.17684v1 Announce Type: new Abstract: Self-evolving agents turn experience into reusable skills, workflows, or memories, but post-evolution accuracy alone does not show whether learned behav
arXiv:2511.19759v3 Announce Type: replace Abstract: Semi-supervised learning (SSL) has emerged as an efficient paradigm for medical image segmentation, reducing the reliance on extensive expert annota
arXiv:2608.16417v1 Announce Type: new Abstract: Multi-modal retrieval-augmented generation (RAG) is a key technique for visually rich long document understanding. Existing multi-modal RAG methods are
arXiv:2608.15447v1 Announce Type: new Abstract: Mobile money has widened financial access across Sub-Saharan Africa and enlarged the surface for money-laundering and terrorism-financing (ML/TF) activi
arXiv:2608.15383v1 Announce Type: new Abstract: Sparse mixture-of-experts (MoE) language models reduce arithmetic by activating only a small subset of experts per token, yet deployment still requires
arXiv:2608.15343v1 Announce Type: new Abstract: Reconstructing three-dimensional computed tomography (CT) from severely constrained projections is highly ill-posed. Sparse angular sampling, restricted
arXiv:2608.15976v1 Announce Type: new Abstract: A learning system can occupy execution states that are indistinguishable under every declared present-behavior readout yet respond differently to future
arXiv:2608.14562v1 Announce Type: new Abstract: AI governance is shifting from voluntary ethics to enforceable, risk-based regulation, yet cross-jurisdictional divergence creates compliance uncertaint
Learn how Jumio built a centralized, real-time feature store on AWS with Amazon SageMaker Feature Store, Amazon Managed Service for Apache Flink, and Amazon Kinesis Data Streams. The architecture deli
arXiv:2608.14656v1 Announce Type: cross Abstract: Photoplethysmography (PPG) is widely used in consumer wearables because of its low cost and ease of acquisition. However, unlike electrocardiography (
arXiv:2608.14650v1 Announce Type: new Abstract: Existing adaptive-inference and world-action-model systems use cheap-stage outputs or predicted futures to allocate additional computation. We study a n
arXiv:2608.16384v1 Announce Type: new Abstract: Universal visual representations require adaptation mechanisms that adapt across heterogeneous domains without fragmenting knowledge into domain-specifi
arXiv:2608.13706v1 Announce Type: cross Abstract: Existing defenses against hallucination in retrieval-augmented and multi-agent pipelines remain partial: evidence is trusted despite modality disagree
arXiv:2608.14425v1 Announce Type: new Abstract: LLM evaluations often use fixed sampling budgets, testing every item the same number of times even after estimates are precise. We introduce optstop, a
arXiv:2608.13546v1 Announce Type: new Abstract: Interactive world models must support persistent memory, responsive interaction, and long-horizon generation, yet these requirements place conflicting d
arXiv:2608.12335v1 Announce Type: new Abstract: Financial question answering over annual reports requires more than retrieving semantically similar passages. It often involves identifying relevant com
arXiv:2608.12626v1 Announce Type: cross Abstract: Strategic reasoning in Large Language Models (LLMs) within long-horizon environments is often limited by inconsistent subgoals. In these settings, fin
arXiv:2608.12435v1 Announce Type: new Abstract: Transformers owe much of their strong long-context retrieval capability to a token-level memory that grows with context length. This flexibility, howeve
arXiv:2608.13112v1 Announce Type: new Abstract: Text-to-image generation has reached photorealistic quality, yet state-of-the-art systems remain unreliable at producing scientific diagrams, whose valu