AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsAgentic Orchestration of EDC Data Space Retrieval and Natural Language Driven Data Processing
arXiv:2610.26548v1 Announce Type: new Abstract: Participating in a data space still requires the Dataspace Protocol (DSP) negotiation flow to be hand-coded against each provider's access policy. We present a multi-agent…
- arXiv cs.AIresearchTraining Natural Schemers: RL Can Amplify Oversight-Conditional Behavior
arXiv:2610.26551v2 Announce Type: replace Abstract: One of the central concerns for the safety of advanced AI is scheming: when a model strategically complies under oversight but pursues its own goals when unobserved. Cu…
- arXiv cs.AIagentsBetaMem: Learning to Remember a Changing World for Embodied Agents
arXiv:2610.26554v1 Announce Type: new Abstract: Long-horizon embodied agents face temporal confusion and incomplete or distorted records, recurring errors from fixed memory rules, and growing storage and retrieval costs.…
- arXiv cs.AIagentsReward Objectives for Robust (PO)MDPs: A Law-Preserving Equivalence with (Partially Observable) Games
arXiv:2610.26568v1 Announce Type: new Abstract: Robust partially observable Markov decision processes model agents that act under partial observability while adversarial environment resolves uncertainty in the transition…
- arXiv cs.AIresearchWhatIf-Bench: Benchmarking Intervention-Aware Time Series Forecasting
arXiv:2610.26582v2 Announce Type: replace Abstract: Recent time series foundation models (TSFMs) can consume covariates in-context, extending zero-shot forecasting to problems driven by external factors. This enables sce…
- arXiv cs.AIagentsWhen the harness decides success: state that cannot be re-observed
arXiv:2610.26632v1 Announce Type: new Abstract: Agent harnesses decide what a language model sees across turns: they drop old turns, summarize them, or replay the model's own earlier reasoning. We ask when these choices…
- arXiv cs.AIresearchRevision-Based Learning from Interactive Experience and Feedback
arXiv:2610.26636v2 Announce Type: replace Abstract: Code generation models are increasingly used to produce interactive artifacts, such as websites and user interfaces. Many properties that determine whether these interf…
- arXiv cs.AIresearchFrom Outcomes to Components: A Counterfactual Evaluation Framework for LLM-Proposed Repairs
arXiv:2610.26650v2 Announce Type: replace Abstract: A repair action proposed by a large language model (LLM) specifies which records to change and how to change their values. A failed action may still contain a useful ch…
- arXiv cs.AIagentsTandem: Serving Hybrid Linear-Attention GUI Agents on Mobile NPUs
arXiv:2610.26660v1 Announce Type: new Abstract: Smartphone GUI agents complete tasks in mobile apps by repeatedly reading the screen and asking a model for the next action. Mobile neural processing units (NPUs) enable lo…
- arXiv cs.AIagentsAllPath: Layered Verification for a Deployed, Self-Revising LLM Trading Agent under Human-Gated and Autonomous Regimes
arXiv:2610.26662v1 Announce Type: new Abstract: LLM trading agents are evaluated almost exclusively in backtests and autonomous benchmark harnesses. Deployment, where actions are irreversible, raises a different question…
- arXiv cs.AIagentsHoneypots for AI Agent Insider Threats
arXiv:2610.26664v1 Announce Type: new Abstract: AI agents deployed inside organizations can pose an insider threat if they make mistakes, are compromised by an attacker, or pursue misaligned objectives. We propose honeyp…
- arXiv cs.AIagentsStepBench: Diagnosing Real-World Procedural Understanding in AI Agents
arXiv:2610.26679v1 Announce Type: new Abstract: Evaluating an agent's planning ability remains challenging for open-domain tasks, where verifying procedures through execution is often impractical. Existing evaluations ty…
- arXiv cs.AIresearchTowards Robust Model Organisms for Eliciting Latent Knowledge
arXiv:2610.26698v1 Announce Type: new Abstract: A key problem in AI safety is eliciting latent knowledge (ELK), extracting information from a model that is misaligned and more knowledgeable than its overseer. For example…
- arXiv cs.AIresearchLG-Bench: A Graph-Structured Evaluation Benchmark for Life Sciences
arXiv:2610.26787v1 Announce Type: new Abstract: Traditional evaluation benchmarks reduce inherently interconnected scientific knowledge in life sciences into flat lists of questions, disregarding the underlying topologic…
- arXiv cs.AIresearchSOMA: Continuous Monitoring and Intervention for System Two Reasoning with System One Models
arXiv:2610.26793v1 Announce Type: new Abstract: Reasoning models often run into problems without noticing them, even when symptoms are visible in their own text. In humans, metacognition separates monitoring from control…
- arXiv cs.AIagentsAgent World Model in the Wild: Computer-Use Agent Environment Synthesis
arXiv:2610.26828v1 Announce Type: new Abstract: Agent learning is shifting from learning from human data toward learning through interaction with the world. However, scaling such learning directly in real-world computer…
- arXiv cs.AIagentsProteinGeoBench: Frontier Structural-Biology Agents Are Fluent at Lookup but Lost in Geometry
arXiv:2610.26829v1 Announce Type: new Abstract: Frontier agents retrieve structural metadata well and compute geometry from coordinates less reliably. We ask how much of a benchmark score reflects that gap and how much r…
- arXiv cs.AIagentsSame Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy
arXiv:2610.26836v1 Announce Type: new Abstract: LLM agents that compute the right answer can still answer wrongly. Agent harnesses increasingly keep long tool outputs out of the context window, saving them to files that…
- arXiv cs.AIagentsAnyGoal: On World Model Generalization Beyond Expert Rollouts
arXiv:2610.26844v2 Announce Type: replace Abstract: The goal of a world model is to capture the dynamics of an environment. This has the great appeal that it allows agents to plan and simulate a course of actions with th…
- arXiv cs.AIagentsTrain An Imaginative Agent
arXiv:2610.26845v1 Announce Type: new Abstract: A seemingly reasonable push in Sokoban can make the puzzle unsolvable. To avoid such mistakes, an agent has to anticipate the consequences of its actions before execution,…
- arXiv cs.AIresearchTransfer Learning for Power Outage Detection Task with Limited Training Data
arXiv:2305.17817v1 Announce Type: cross Abstract: Early detection of power outages is crucial for maintaining a reliable power distribution system. This research investigates the use of transfer learning and language mod…
- arXiv cs.AIresearchMulti-LLM Systems Exhibit Robust Semantic Collapse
arXiv:2605.17193v3 Announce Type: cross Abstract: Whether machines can originate novel content has been debated for nearly two centuries, from Lovelace's assertion that no engine can "originate anything" to Turing's ques…
- arXiv cs.AIresearchSAMPA: Transformer-based segmentation of prosodic boundaries in Brazilian Portuguese
arXiv:2607.07408v3 Announce Type: replace-cross Abstract: Automatic prosodic segmentation identifies boundaries between speech units based on acoustic and linguistic evidence. Although recent deep learning approaches hav…
- arXiv cs.AIresearchCan the Quantum Fisher Effective Dimension Judge a Hybrid Quantum Neural Network?
arXiv:2610.14070v1 Announce Type: cross Abstract: Hybrid quantum neural networks (QNNs) pair a classical feature extractor with a variational quantum circuit, and the effective dimension of the circuit's quantum Fisher i…
- arXiv cs.AIresearchAnytime Contracts for AI Services
arXiv:2610.22781v1 Announce Type: cross Abstract: AI services are prone to moral hazard, as model providers can silently route requests to cheaper, lower-quality models, and the inherent stochasticity of model outputs ma…
- arXiv cs.AIresearchLLM-Assisted Bi-Approximate CHC Solving
arXiv:2610.23771v1 Announce Type: cross Abstract: Constrained Horn clause (CHC) solvers often struggle to discover deep counterexamples whose derivations may require extensive rule unfolding. Although large language mode…
- arXiv cs.AIagentsWeb Agent Agentic Reinforcement Learning Decision Model Under Multi-Cost and Failure Risk
arXiv:2610.23857v1 Announce Type: cross Abstract: Intelligent agent interactions in real-world web environments are commonly constrained by request budgets, time delays, anti-crawling restrictions, and operational failur…
- arXiv cs.AIresearchComposable Parametric Demographic Injection for Values Survey Simulation
arXiv:2610.23867v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as simulated respondents in computational social science, yet accurately reproducing population-specific heterogeneous…
- arXiv cs.AIagentsContext-Aware RGB-Thermal-Radar Obstacle Perception for Resource-Constrained ASVs
arXiv:2610.23873v1 Announce Type: cross Abstract: Reliable obstacle perception is challenging for small autonomous surface vessels (ASVs): reflections and glare confound visible-light cameras, vessel motion shifts the ap…
- arXiv cs.AIagentsLatent Multimodal Memory Reasoning
arXiv:2610.23890v1 Announce Type: cross Abstract: Long-term personalized agents are required to reason over distributed history across modalities and evolving dialogue states. While existing flat cross-modal methods comp…
- arXiv cs.AIagentsEMMA: Event-Driven Multimodal Memory System with Multi-Agent Retrieval
arXiv:2610.23892v1 Announce Type: cross Abstract: Long video understanding requires reasoning over hours of content that far exceed the context window of multimodal large language models (MLLMs). Existing approaches comp…
- arXiv cs.AIresearchGeoForward: Geometry-Guided Cross-View Alignment for Feedforward 3D Gaussian Splatting
arXiv:2610.23895v1 Announce Type: cross Abstract: Feedforward 3D reconstruction has largely followed an {appearance-driven paradigm}: multi-view images are encoded into visual representations, cross-view correspondence i…
- arXiv cs.AIresearchEgoVine: Post-Training with Branching Experience Compiled from Real-World Videos
arXiv:2610.23907v1 Announce Type: cross Abstract: Real-world egocentric videos capture multi-step household procedures, but their recorded trajectories leave alternative actions and recovery from the learner's own mistak…
- arXiv cs.AIresearchWhen the Editing Intent Is Split: A Cross-Modal Jailbreak Attack for Large Image Editing Models
arXiv:2610.23909v1 Announce Type: cross Abstract: Large image editing models increasingly infer editing intent from both textual and visual inputs, creating an underexplored safety risk when harmful intent is distributed…
- arXiv cs.AIresearchUMI-WM: Interactive World Modeling for Bimanual Manipulation from Wrist Cameras
arXiv:2610.23913v1 Announce Type: cross Abstract: Universal Manipulation Interface (UMI) pairs wrist-camera videos with camera trajectories and gripper openings, providing scalable, robot-free data for learning how biman…
- arXiv cs.AIresearchSPIRIT: Sparse Physics-Informed Inverse Rendering for SAR Imaging
arXiv:2610.23914v1 Announce Type: cross Abstract: Radio-frequency (RF) and wave-based imaging systems play a critical role in sensing under adverse conditions, but are fundamentally constrained by a trade-off between col…
- arXiv cs.AIresearchCross-Modal Forgetting and Transfer in Unified Multimodal Post-Training
arXiv:2610.23918v1 Announce Type: cross Abstract: Unified multimodal models answer questions about images and also generate images with one shared set of weights. This creates a post-training risk that single-purpose mod…
- arXiv cs.AIresearchZero-Shot Tabular Learning via Hypergraph Reasoning with LLMs
arXiv:2610.23923v1 Announce Type: cross Abstract: Tabular data are prevalent across diverse domains such as healthcare and finance. Zero-shot tabular learning aims to make predictions without labeled data. However, exist…
- arXiv cs.AIresearchSpinDoctor: Finding Innocuous Prompts for Biased Image Generation
arXiv:2610.23925v1 Announce Type: cross Abstract: We introduce SpinDoctor, a method for discovering innocuous prompts for text-to-image models that result in images with a chosen target bias, while retaining the quality…
- arXiv cs.AIresearchImproving Fake Profile Detection Through Multimodal Learning and Research-Guided Feature Integration
arXiv:2610.23926v1 Announce Type: cross Abstract: Fake accounts continue to undermine online platforms, yet it remains unclear whether multimodal learning offers meaningful improvements over strong metadata baselines or…
- arXiv cs.AIresearchminiTE: Diagnosing and Bridging the Execution Gap in Native NVFP4 Reinforcement Learning
arXiv:2610.23928v1 Announce Type: cross Abstract: Native NVFP4 promises to accelerate reinforcement learning (RL) for large language models (LLMs), but achieving stable learning across separate training and sampling engi…
- arXiv cs.AIresearchOrganizing LLM Inference for Aggregate Predictions from Behavioral Data
arXiv:2610.23934v1 Announce Type: cross Abstract: When using a large language model (LLM) to simulate a user population, how should we combine individual behavioral histories to predict the fraction of users who take a p…
- arXiv cs.AIresearchFactorization-Guided Training and Inference for Diffusion-Drafted Speculative Decoding
arXiv:2610.23935v1 Announce Type: cross Abstract: Speculative decoding accelerates large language model inference through parallel verification of tokens proposed by a drafter. Diffusion drafters further parallelize prop…
- arXiv cs.AIresearchBeyond Prediction: A Component-Level Study of Concept Recovery in In-Context Learning
arXiv:2610.23936v1 Announce Type: cross Abstract: In-context learning (ICL) enables language models to make accurate predictions from demonstrations, but prediction success alone does not reveal the extent to which the u…
- arXiv cs.AIresearchSECURE-AI: Multi-Signal Uncertainty Quantification for Trustworthy Acoustic Cough Detection
arXiv:2610.23937v1 Announce Type: cross Abstract: Acoustic models for cough detection achieve high accuracy on curated benchmarks but fail silently in deployment, where speech, environmental noise, and audio unlike the t…
- arXiv cs.AIresearchPeekKV: The First Content Token Already Knows Which Heads Matter
arXiv:2610.23939v1 Announce Type: cross Abstract: The size of the KV cache grows linearly with context length, making it a major memory bottleneck in LLM serving. Motivated by recent progress in identifying the functiona…
- arXiv cs.AIresearchA Group-Aware Comparative Study of Classical Machine Learning and Deep Learning Methods for Sensor-Based Human Activity Recognition
arXiv:2610.23941v1 Announce Type: cross Abstract: Human Activity Recognition (HAR) using wearable or smartphone inertial sensors is commonly evaluated with both classical machine learning and deep learning methods. Howev…
- arXiv cs.AIagentsReTTT: Reflection-Driven Test-Time Training via Experience-Based Parametric Memory
arXiv:2610.23942v1 Announce Type: cross Abstract: LLM agents accumulate valuable task-specific information through their actions, observations, and environmental feedback during multi-turn task execution. When effectivel…
- arXiv cs.AIresearchCrowdH-Edit: Reference-Guided Instruction-Based Synthetic Hajj Crowd Image Generation for Crowd Counting
arXiv:2610.23945v1 Announce Type: cross Abstract: The scarcity of domain-specific annotated images remains a major limitation for developing computer-vision systems for Hajj crowd analysis. Existing crowd-counting datase…
- arXiv cs.AIresearchREAP: Learning General Rearrangement Heuristics from Motion-Planner Feedback
arXiv:2610.23973v1 Announce Type: cross Abstract: Robotic rearrangement tasks require planning sequences of actions that fulfill high-level specifications while also being geometrically feasible. Symbolic abstractions ma…