AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchSpatial disparities in simulated patrol detection: a reproducible generative-model audit across Baltimore and Chicago
arXiv:2603.18987v3 Announce Type: replace Abstract: Auditing predictive policing requires distinguishing recorded crime incidents, simulated patrol observations, and the demographic characteristics of places. We present…
- arXiv cs.AIresearchUncertainty-Guided Latent Diagnostic Trajectory Learning for Sequential Clinical Diagnosis
arXiv:2604.05116v2 Announce Type: replace Abstract: Clinical diagnosis requires sequential evidence acquisition under uncertainty. However, most Large Language Model (LLM) based diagnostic systems assume fully observed p…
- arXiv cs.AIresearchNoise-Robust Financial Numerical Entity Attribute Tagging
arXiv:2605.24910v2 Announce Type: replace Abstract: Financial Numerical Entity (FNE) understanding aims to recover the meaning of numerical mentions in financial reports. Existing studies focus on concept name prediction…
- arXiv cs.AIresearchCan Broad Biomedical Knowledge be Contextualized into Scenario-Grounded Propositions?
arXiv:2605.27082v2 Announce Type: replace Abstract: Biomedical discovery often requires connecting broad biomedical knowledge with specific experimental or clinical data. Background knowledge suggests relevant mechanisms…
- arXiv cs.AIresearchSame Answer, Different Confidence: Protocol Sensitivity in LLM Confidence Calibration
arXiv:2605.27752v4 Announce Type: replace Abstract: Large language models (LLM) can state their confidence in an answer (verbalized confidence), or the confidence can be read from the probability of the answer's tokens (…
- arXiv cs.AIresearchTrust the Right Teacher: Quality-Aware Self-Distillation for GUI Grounding
arXiv:2606.18101v3 Announce Type: replace Abstract: GUI grounding requires vision-language models (VLMs) to identify small target elements in high-resolution screenshots and predict precise screen coordinates. On-policy…
- arXiv cs.AIresearchDeFAb: A Dataset for Grounded Scientific Discovery through Defeasible Abduction
arXiv:2606.18557v2 Announce Type: replace Abstract: A theory can accommodate a counterexample by adding an exception. The harder problem is deciding how far the exception should extend and which predictions should surviv…
- arXiv cs.AIresearchAttractor Domain Theory: A Mathematical Framework for Cardiovascular Attractor Analysis with Wearable Photoplethysmography (PPG) Validation
arXiv:2606.22039v3 Announce Type: replace Abstract: The cardiac attractor provides a phase-space representation of cardiovascular dynamics reconstructed from scalar observations under standard delay-embedding conditions.…
- arXiv cs.AIresearchDeepSafe-SciHazard: A Benchmark for Measuring Scientific Safety Risks with Decomposed Harm Scoring
arXiv:2607.18665v2 Announce Type: replace Abstract: Large language models (LLMs) increasingly support science, but they can also convert hazardous scientific knowledge into actionable misuse guidance. Existing benchmarks…
- arXiv cs.AIresearchWhere Cross-Question Structure in LLM Errors Comes From: Shared Seeds, Shared Prompts, and Answer-Letter Preference
arXiv:2607.20464v2 Announce Type: replace Abstract: Disagreement between language models is often read as missing knowledge. Questions that need the same missing knowledge should then be right or wrong together. We test…
- arXiv cs.AIresearchShapes from Examples: Foundations of Shape Learning in Recursive SHACL
arXiv:2607.27934v3 Announce Type: replace Abstract: SHACL shapes enable data graph validation, making automatic shape learning essential for knowledge graph applications. We investigate the well-known fitting approach to…
- arXiv cs.AIresearchAI-guided deep visual proteomics reveals recurrence-associated morpho-molecular signatures within triple-negative breast cancers
arXiv:2608.03145v3 Announce Type: replace Abstract: Outcome informed AI can localize prognostically relevant regions in digital pathology, but directly characterizing the molecular basis of these spatial predictions rema…
- arXiv cs.AIresearchTreeWY: Speculative Verification for Gated DeltaNet Hybrids
arXiv:2608.20961v2 Announce Type: replace Abstract: Modern open models are hybrids: most layers are linear-attention (Gated DeltaNet, GDN) layers carrying a small fixed-size recurrent state instead of a growing key-value…
- arXiv cs.AIresearchOn Orthogonal Recurrent Neural Networks Quantization
arXiv:2402.04012v3 Announce Type: replace-cross Abstract: In recent years, Orthogonal Recurrent Neural Networks (ORNNs) have gained popularity due to their ability to manage tasks involving long-term dependencies, such a…
- arXiv cs.AIresearchPower- and Fragmentation-aware Online Scheduling for GPU Datacenters
arXiv:2412.17484v2 Announce Type: replace-cross Abstract: The rise of Artificial Intelligence and Large Language Models is driving increased GPU usage in data centers for complex training and inference tasks, impacting o…
- arXiv cs.AIresearchEnhancing Table Recognition with Vision LLMs: A Benchmark and Neighbor-Guided Toolchain Reasoner
arXiv:2412.20662v4 Announce Type: replace-cross Abstract: Pre-trained foundation models have recently made significant progress in table-related tasks such as table understanding and reasoning. However, recognizing the s…
- arXiv cs.AIresearchCatastrophic Overfitting, Entropy Gap and Participation Ratio: A Noiseless $l^p$ Norm Solution for Fast Adversarial Training
arXiv:2505.02360v3 Announce Type: replace-cross Abstract: Adversarial training is a cornerstone of robust deep learning, but fast methods like the Fast Gradient Sign Method (FGSM) often suffer from Catastrophic Overfitti…
- arXiv cs.AIresearchDeep Learning Warm Starts for Trajectory Optimization on the International Space Station
arXiv:2505.05588v4 Announce Type: replace-cross Abstract: Trajectory optimization is a cornerstone of modern robot autonomy, enabling systems to compute trajectories and controls in real-time while respecting safety and…
- arXiv cs.AIresearchAU-Harness: An Open-Source Toolkit for Efficient and Unified Evaluation of AUdio-LLMs
arXiv:2509.08031v4 Announce Type: replace-cross Abstract: Large Audio Language Models (LALMs) are rapidly advancing, but evaluating them remains challenging due to inefficient and nonstandardized toolkits that limit fair…
- arXiv cs.AIresearchDynamic Relational Priming Improves Transformer in Multivariate Time Series
arXiv:2509.12196v3 Announce Type: replace-cross Abstract: Standard attention mechanisms in transformers employ static token representations that remain unchanged across all pair-wise computations in each layer. This limi…
- arXiv cs.AIresearchHuman-AI Interactions and the Psychological Triad: A Multi-Domain Survey of Cognitive, Behavioral, and Emotional Impacts
arXiv:2510.17753v3 Announce Type: replace-cross Abstract: As human-AI interactions increasingly reshape social and research landscapes, the risks of overreliance, cognitive offloading, and the nuanced degradation of huma…
- arXiv cs.AIresearchBeyond the Monolithic Scheduler: Lightweight Adaptive Policy Switcher for Process Scheduling Optimization
arXiv:2511.11628v2 Announce Type: replace-cross Abstract: Modern operating system process schedulers typically enforce a static, ``one-policy-fits-all'' design, which inevitably compromises performance under the highly d…
- arXiv cs.AIresearchEvalCards: A Framework for Standardized Evaluation Reporting
arXiv:2511.21695v2 Announce Type: replace-cross Abstract: Evaluation reports guide the use of general-purpose AI (GPAI) models, yet inconsistent reporting often leaves essential questions unanswered. Through a review of…
- arXiv cs.AIresearchDementia-R1: Reinforced Pretraining and Reasoning from Unstructured Clinical Notes for Real-World Dementia Prognosis
arXiv:2601.03018v3 Announce Type: replace-cross Abstract: While Large Language Models (LLMs) have shown strong performance on clinical text understanding, they struggle with longitudinal prediction tasks such as dementia…
- arXiv cs.AIresearchThe Role of Quantum in Hybrid Quantum-Classical Neural Networks: An Empirical Assessment
arXiv:2601.04732v2 Announce Type: replace-cross Abstract: Quantum machine learning has emerged as a promising application domain for near-term quantum hardware, particularly through hybrid quantum-classical models that l…
- arXiv cs.AIresearchAttention-space Contrastive Guidance for Efficient Hallucination Mitigation in LVLMs
arXiv:2601.13707v3 Announce Type: replace-cross Abstract: Hallucinations in large vision--language models (LVLMs) often arise when language priors dominate over visual evidence, leading to object misidentification and vi…
- arXiv cs.AIresearchTowards robust long-context understanding of large language model via active recap learning
arXiv:2601.13734v2 Announce Type: replace-cross Abstract: In this paper, we propose active recap learning (ARL), a framework for enhancing large language model (LLM) in understanding long contexts. ARL enables models to…
- arXiv cs.AIresearchOmniSpectra: A Unified Foundation Model for Universal Spectra Representation Learning
arXiv:2601.15351v2 Announce Type: replace-cross Abstract: Standard sequence models encode position by token rank rather than physical coordinate, typically forcing practitioners to destructively resample continuous physi…
- arXiv cs.AIresearchClifford Kolmogorov-Arnold Networks
arXiv:2602.05977v3 Announce Type: replace-cross Abstract: We introduce Clifford Kolmogorov-Arnold Network (ClKAN), a flexible and efficient architecture for function approximation in arbitrary Clifford Algebra spaces. We…
- arXiv cs.AIresearchStealthRL: Reinforcement Learning Paraphrase Attacks for Multi-Detector Evasion of AI-Text Detectors
arXiv:2602.08934v3 Announce Type: replace-cross Abstract: AI-text detectors are increasingly used in high-stakes settings, but their robustness to meaning-preserving adversarial rewriting remains unclear. We introduce St…
- arXiv cs.AIresearchMitigating the Likelihood Paradox in Flow-based OOD Detection: An Entropic Perspective
arXiv:2602.09581v3 Announce Type: replace-cross Abstract: Deep generative models that can tractably compute input likelihoods, including normalizing flows, often assign unexpectedly high likelihoods to out-of-distributio…
- arXiv cs.AIresearchWhy Does Counterintuitive Likelihood Behavior Rarely Emerge in Tabular Anomaly Detection?
arXiv:2602.09593v3 Announce Type: replace-cross Abstract: Deep generative models with tractable likelihoods, exemplified by normalizing flows, offer an effective basis for anomaly detection through likelihood-based scori…
- arXiv cs.AIresearchCrimeNER: Curated Dataset and Methods for Research and Investigation in Language Intelligence
arXiv:2603.02150v3 Announce Type: replace-cross Abstract: The extraction of critical information from crime-related documents is a crucial task for law enforcement agencies. The extraction of this information can be inte…
- arXiv cs.AIresearchVORL-EXPLORE: A Hybrid Learning Planning Approach to Multi-Robot Exploration in Dynamic Environments
arXiv:2603.07973v2 Announce Type: replace-cross Abstract: Hierarchical multi-robot exploration decouples frontier allocation from local navigation, making systems brittle in dynamic environments. Without awareness of exe…
- arXiv cs.AIresearchMoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenarios
arXiv:2603.09983v2 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) models enable scalable performance but face severe memory constraints on edge devices. Existing offloading strategies struggle with I/O b…
- arXiv cs.AIresearchMetaKE: Meta-Learning for Knowledge Editing Toward a Better Accuracy-Editability Trade-off
arXiv:2603.12677v4 Announce Type: replace-cross Abstract: Existing locate-then-edit Knowledge Editing (KE) methods typically decompose editing into two stages: upstream target representation optimization and downstream c…
- arXiv cs.AIresearchBetter Reward Models Are Not Always Less Biased: Social Alignment Beyond Standard Reward Benchmarks
arXiv:2605.05003v2 Announce Type: replace-cross Abstract: Reward models are central to LLM alignment, yet current benchmarks reveal little about how they represent diverse and sometimes competing social preferences. We e…
- arXiv cs.AIresearchMoLS: SNR-Guided Module-Wise Learning-Rate Calibration for Large Language Models
arXiv:2605.05794v2 Announce Type: replace-cross Abstract: Large language models (LLMs) contain functionally distinct modules with heterogeneous gradient statistics. Although Adam(W) adapts updates at the parameter level,…
- arXiv cs.AIresearchQ-MMR: Off-Policy Evaluation via Recursive Reweighting and Moment Matching
arXiv:2605.06474v3 Announce Type: replace-cross Abstract: We present a novel theoretical framework, Q-MMR, for off-policy evaluation in finite-horizon MDPs. Q-MMR learns a set of scalar weights, one for each data point,…
- arXiv cs.AIresearchMedAction: Towards Active Multi-turn Clinical Diagnostic LLMs
arXiv:2605.07305v2 Announce Type: replace-cross Abstract: Most existing LLM diagnoses are evaluated on static, single-turn settings where complete patient information is provided upfront, an oversimplification of real cl…
- arXiv cs.AIresearchPOETS: Uncertainty-Aware LLM Optimization via Compute-Efficient Policy Ensembles
arXiv:2605.07775v2 Announce Type: replace-cross Abstract: Balancing exploration with exploitation is a core challenge in black-box optimization and reinforcement learning. We introduce POETS ($\textbf{Po}$licy $\textbf{E…
- arXiv cs.AIresearchAttractor-Vascular Coupling Theory: Formal Grounding and Empirical Validation for AAMI-Standard Cuffless Blood Pressure Estimation from Smartphone Photoplethysmography
arXiv:2605.10871v3 Announce Type: replace-cross Abstract: We present Attractor-Vascular Coupling Theory (AVCT), a theory-guided framework for deriving interpretable blood pressure (BP) predictors from photoplethysmograph…
- arXiv cs.AIresearchVLMs Trace Without Tracking: Diagnosing Failures in Visual Path Following
arXiv:2605.15672v2 Announce Type: replace-cross Abstract: Vision-language models (VLMs) perform strongly on multimodal benchmarks, yet their ability to reliably follow continuous visual structure remains unclear. We focu…
- arXiv cs.AIresearchShaping the Prior: How Synthetic Task Distributions Determine Tabular Foundation Model Quality
arXiv:2605.18971v2 Announce Type: replace-cross Abstract: What determines the quality of a tabular foundation model? Unlike language or vision, tabular foundation models acquire their inductive biases almost entirely fro…
- arXiv cs.AIresearchEfficient Benchmarking Is Just Feature Selection and Multiple Regression
arXiv:2605.25773v3 Announce Type: replace-cross Abstract: Efficient benchmarking techniques aim to lower the computational cost of evaluating LLMs by predicting full benchmark scores using only a subset of a benchmark's…
- arXiv cs.AIresearchProbing Collision Grounding in Vision-Language Models for Safe Human-Robot Collaboration
arXiv:2605.31196v2 Announce Type: replace-cross Abstract: Safe human--robot collaboration requires more than visual description: a monitor must determine whether the robot body is safely separated, already colliding with…
- arXiv cs.AIresearchStemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning
arXiv:2606.00148v2 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) often know the rule but pick the wrong answer: on abstract visual reasoning (AVR) tasks, a model can describe what it see…
- arXiv cs.AIresearchContinuous Audio Thinking for Large Audio Language Models
arXiv:2606.18273v2 Announce Type: replace-cross Abstract: Large audio language models (LALMs) have shown impressive capabilities on diverse audio understanding tasks, ranging from speech transcription to music analysis.…
- arXiv cs.AIresearchSteerVTE: Seamless Video Text Editing with Style and Glyph Control
arXiv:2606.23254v2 Announce Type: replace-cross Abstract: Visual text editing aims to precisely modify text in images and videos while preserving stylistic consistency and visual realism. Despite significant advances in…
- arXiv cs.AIresearchWorld from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video
arXiv:2607.01202v2 Announce Type: replace-cross Abstract: We present World from Motion, a method for generating freely renderable dynamic 3D Gaussian representations from monocular videos. Our approach conditions a video…