AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchUxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence
arXiv:2605.09040v4 Announce Type: replace Abstract: Modeling ultra-long user sequences involves a difficult trade-off between efficiency and effectiveness. While current paradigms rely on either item-specific search or i…
- arXiv cs.AIagentsWhen Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents
arXiv:2605.11928v2 Announce Type: replace Abstract: Tool-use language agents are evaluated on benchmarks that assume clean inputs, unambiguous tool registries, and reliable APIs. Real deployments violate all these assump…
- arXiv cs.AIresearchFrom SGD to Muon: Adaptive Optimization via Schatten-p Norms
arXiv:2605.19781v3 Announce Type: replace Abstract: Modern optimizers, like Muon, impose matrix-wise geometry constraints on their updates. These matrix-wise constraints can be unified under Linear Minimization Oracle (L…
- arXiv cs.AIresearchMPMMine: Better Benchmarks for Constraint Acquisition
arXiv:2605.26279v2 Announce Type: replace Abstract: Constraint Acquisition (CA) and related research on the validation and enhancement of Mathematical Programming (MP) models using domain knowledge artifacts are currentl…
- arXiv cs.AIresearchLatent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs
arXiv:2606.00726v4 Announce Type: replace Abstract: Strong reasoning depends not only on model knowledge but also on how effectively cognitive behaviors are deployed during generation. Existing methods often rely on expl…
- arXiv cs.AIagentsEvoDrive: Pareto Evolution for Safety-Critical Autonomous Driving via Self-Improving LLM Agents
arXiv:2606.03678v2 Announce Type: replace Abstract: Generating safety-critical scenarios is essential for validating and improving autonomous driving systems, yet it inherently requires maximizing adversariality to expos…
- arXiv cs.AIagentsThe Saturation Trap and the Subjectivity of Intervention Timing: Why Affect-Based Triggers and LLM Judges Fail to Time Interventions on Autonomous Agents
arXiv:2606.04296v2 Announce Type: replace Abstract: As autonomous AI agents move from conversational systems to long-horizon software execution, runtime safety layers that decide when to interrupt an agent have become es…
- arXiv cs.AIagentsAgents' Last Exam
arXiv:2606.05405v4 Announce Type: replace Abstract: Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many pr…
- arXiv cs.AIagentsDo More Agents Help? Evaluating Accuracy and Cost in LLM Multi-Agent Workflows
arXiv:2606.05670v2 Announce Type: replace Abstract: Choosing whether to replace a single-agent controller requires evidence on both task performance and inference expenditure. We present BenchAgent, a shared evaluation f…
- arXiv cs.AIagentsExperience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory
arXiv:2606.09365v4 Announce Type: replace Abstract: Medical agent systems are increasingly expected to support interactive clinical decision making rather than only static question answering. In such settings, effective…
- arXiv cs.AIagentsTelcoAgent: A Scalable 5G Multi-KPM Forecasting With 3GPP-Grounded Explainability
arXiv:2606.19821v2 Announce Type: replace Abstract: Key Performance Measurement (KPM) forecasting is essential for proactive network management of 5G and next-generation telecom networks. However, existing machine learni…
- arXiv cs.AIagentsHierarchical Experimentalist Agents
arXiv:2606.29315v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used to act in the real world and augment human decision-making, yet most systems rely on parametric knowledge acquired th…
- arXiv cs.AIagentsAgent Step Value: Auditing Evaluation-Channel Sensitivity in Black-Box Agent Traces
arXiv:2607.04419v5 Announce Type: replace Abstract: Reusing evaluator-derived step rewards across evaluation channels raises a measurement question: does the assigned direction remain stable? We define Agent Step Value (…
- arXiv cs.AIagentsMeasuring Harness-Induced Belief Divergence in Multi-Step LLM Agents
arXiv:2607.04528v3 Announce Type: replace Abstract: Software agents are becoming increasingly important in modern coding workflows, where they solve complex tasks through multi-step interaction with code repositories. A…
- arXiv cs.AIagentsWho&When Pro: Can LLMs Really Attribute Failures in AI Agents?
arXiv:2607.09996v2 Announce Type: replace Abstract: Automated failure attribution uses LLMs to identify where and why agentic systems fail. As agents become more capable, their failures become subtler, making automated a…
- arXiv cs.AIresearchMIRROR: Learning from the Other View for Multi-Modal Reasoning
arXiv:2607.21552v2 Announce Type: replace Abstract: Vision-language models (VLMs) can behave very differently depending on whether the same underlying problem is presented textually or visually. This inconsistency sugges…
- arXiv cs.AIresearchEnhancing Tabular Learners with Context-Aware Semantic Embeddings
arXiv:2608.03565v2 Announce Type: replace Abstract: While modern tabular learners excel at capturing statistical patterns, they frequently operate in a semantic vacuum, treating textual features as discrete symbols, igno…
- arXiv cs.AIagentsCASCADE: An Agentic Regulatory Network Framework for Patient-Data-Validated Downstream Perturbation Prediction
arXiv:2608.05359v3 Announce Type: replace Abstract: CASCADE is an agentic framework that predicts downstream transcriptional effects of gene perturbation from precomputed ARACNe regulatory networks, exposed via MCP. Prio…
- arXiv cs.AIresearchResidencyRL: Reinforcement Learning in Simulated Clinical Environments
arXiv:2608.07418v2 Announce Type: replace Abstract: In medical education, physicians convert academic knowledge into clinical expertise through residency: years of training across thousands of encounters, with diverse so…
- arXiv cs.AIresearchDifferent Feedback, Different Updates: Selective Self-Learning from User Interactions for Large Language Models
arXiv:2608.09109v2 Announce Type: replace Abstract: User feedback offers natural supervision for persistent LLM improvement, but a single message may support multiple behavioral changes with different scopes of generaliz…
- arXiv cs.AIresearchFrom Mastery Profile to Simulated Response: Stochastic Student Knowledge Graphs (SSKG) for Faithful LLM Student Simulation
arXiv:2608.21668v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used to simulate students at different mastery levels to generate synthetic training data and stress-test tutoring systems…
- arXiv cs.AIagentsGenCoord: Executable Commitments for Embodied LLM Agents under Private Information
arXiv:2608.22055v3 Announce Type: replace Abstract: Embodied language-model agents must coordinate interdependent actions while each holds private task information. In dialogue-based coordination, models generate message…
- arXiv cs.AIresearchPreference Elicitation for Policy Optimization and Application to Aligning Heart Transplantation with Human Values
arXiv:2608.28620v2 Announce Type: replace Abstract: Preference elicitation is essential for aligning AI systems with human values. Prior approaches (e.g., for organ allocation) often ask stakeholders to compare the decis…
- arXiv cs.AIagentsCan escalation channels redirect reward hacking toward defect disclosure?
arXiv:2608.29460v3 Announce Type: replace Abstract: When coding agents encounter defective test infrastructure they may reward-hack: hardcoding outputs or editing test files to pass tests they cannot legitimately satisfy…
- arXiv cs.AIagentsData-Driven Persona-Conditioned Agents for A/B Test Simulation
arXiv:2609.01038v2 Announce Type: replace Abstract: A/B testing is the gold standard for evaluating product changes, but each experiment requires real user traffic, engineering effort, and weeks of measurement. We propos…
- arXiv cs.AIresearchH2Table: Hierarchical Hypergraph-Enhanced Large Language Models for Complex Table Reasoning
arXiv:2609.01216v2 Announce Type: replace Abstract: Tables are ubiquitous across diverse domains, yet reasoning over them remains a significant challenge for modern large language models (LLMs). Current approaches typica…
- arXiv cs.AIagentsMulti-Harness Reinforcement Learning for Coding Agents: Training Gains and Harness Dependence
arXiv:2609.04518v2 Announce Type: replace Abstract: Recent systems train coding agents with reinforcement learning (RL) on trajectories collected through multiple execution harnesses. However, gains under the training ha…
- arXiv cs.AIresearchAPI Benchmark Scores Do Not Reliably Transfer to Chatbot Interfaces
arXiv:2609.08861v3 Announce Type: replace Abstract: Benchmark scores are a central currency in model releases: they inform purchasing decisions, shape public trust, and influence policy. Yet, a key assumption underlying…
- arXiv cs.AIresearchProvably Accurate Shapley Value Estimation via Leverage Score Sampling
arXiv:2410.01917v3 Announce Type: replace-cross Abstract: Originally introduced in game theory, Shapley values have emerged as a central tool in explainable machine learning, where they are used to attribute model predic…
- arXiv cs.AIresearchConditional Vendi Score: Prompt-Aware Diversity Evaluation for Generative AI Models and LLMs
arXiv:2411.02817v3 Announce Type: replace-cross Abstract: Generative models guided by text prompts are widely evaluated for fidelity and prompt alignment, yet their ability to produce diverse outputs remains underexplore…
- arXiv cs.AIresearchUnderstanding Transformer-based Vision Models through Inversion
arXiv:2412.06534v5 Announce Type: replace-cross Abstract: Understanding the internal mechanisms of deep neural networks remains a central challenge in machine learning. In computer vision, one promising yet only prelimin…
- arXiv cs.AIresearchCRM: Retrieval Model with Controllable Condition
arXiv:2412.13844v2 Announce Type: replace-cross Abstract: Retrieval models for recommendation need to account for outcomes such as watch time while searching large item corpora efficiently. We present the Controllable Re…
- arXiv cs.AIresearchWoundAmbit: Bridging State-of-the-Art Semantic Segmentation and Real-World Wound Care
arXiv:2504.06185v3 Announce Type: replace-cross Abstract: Chronic wounds affect a large population, particularly the elderly and diabetic patients, who often exhibit limited mobility and co-existing health conditions. Au…
- arXiv cs.AIresearchHeterogeneity-Aware Client Sampling for Optimal and Efficient Federated Learning
arXiv:2505.11304v4 Announce Type: replace-cross Abstract: Federated learning (FL) commonly involves clients with diverse communication and computational capabilities. Such heterogeneity can significantly distort the opti…
- arXiv cs.AIresearchAuditing Prompt Injection Defenses Before They Reach Users: What Strong Adversaries Reveal
arXiv:2505.15738v3 Announce Type: replace-cross Abstract: Large language models are increasingly deployed in user-facing products that handle sensitive personal data while interacting with untrusted external content. Thi…
- arXiv cs.AIresearchReasoningShield: Safety Detection over Reasoning Traces of Large Reasoning Models
arXiv:2505.17244v3 Announce Type: replace-cross Abstract: Large Reasoning Models (LRMs) leverage explicit reasoning traces, known as Chain-of-Thought (CoT) reasoning, to decompose complex problems into intermediate steps…
- arXiv cs.AIresearchBreaking the Cloak! Unveiling Chinese Cloaked Toxicity with Homophone Graph and Toxic Lexicon
arXiv:2505.22184v3 Announce Type: replace-cross Abstract: Social media platforms have experienced a significant rise in toxic content, including abusive language and discriminatory remarks, presenting growing challenges…
- arXiv cs.AIresearchPhysics-inspired Generative AI models via real hardware-based noisy quantum diffusion
arXiv:2505.22193v2 Announce Type: replace-cross Abstract: Quantum Diffusion Models (QDMs) are an emerging paradigm in Generative AI that aims to use quantum properties to improve the performances of their classical count…
- arXiv cs.AIresearchUniConFlow: A Unified Constrained Flow-Matching Framework for Certified Motion Planning
arXiv:2506.02955v3 Announce Type: replace-cross Abstract: Generative models have become increasingly powerful tools for robot motion generation, enabling flexible and multimodal trajectory generation across various tasks…
- arXiv cs.AIresearchLLMail-Inject: A Dataset from a Realistic Adaptive Prompt Injection Challenge
arXiv:2506.09956v2 Announce Type: replace-cross Abstract: Indirect prompt injection attacks exploit a fundamental weakness of LLMs: executing unauthorized tasks from untrusted sources. Despite the practical significance…
- arXiv cs.AIresearchRegression-adjusted Monte Carlo Estimators for Shapley Values and Probabilistic Values
arXiv:2506.11849v3 Announce Type: replace-cross Abstract: With origins in game theory, probabilistic values like Shapley values, Banzhaf values, and semi-values have emerged as a central tool in explainable AI. They are…
- arXiv cs.AIresearchTowards a Measure Theory of Semantic Information
arXiv:2508.00525v2 Announce Type: replace-cross Abstract: A classic account of the quantification of semantic information is that of Bar-Hiller and Carnap. Their account proposes an inverse relation between the informati…
- arXiv cs.AIresearchAccess Paths for Efficient Ordering with Large Language Models
arXiv:2509.00303v5 Announce Type: replace-cross Abstract: In this work, we present the \texttt{LLM ORDER BY} semantic operator as a logical abstraction and conduct a systematic study of its physical implementations. Firs…
- arXiv cs.AIresearchDiscovering Phase Space Structure in Learned Hamiltonian Systems
arXiv:2509.23003v3 Announce Type: replace-cross Abstract: Mechanics-inspired neural dynamical models improve interpretability and physical consistency, but they usually assume a known configuration space, fixed physical…
- arXiv cs.AIresearchABLEIST: Intersectional Disability Bias in LLM-Generated Hiring Scenarios
arXiv:2510.10998v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly under scrutiny for perpetuating identity-based discrimination in high-stakes domains such as hiring, particularly ag…
- arXiv cs.AIresearchA Theory of the Mechanics of Information: Generalization Through Measurement of Uncertainty (Learning is Measuring)
arXiv:2510.22809v2 Announce Type: replace-cross Abstract: Traditional machine learning relies on explicit models and domain assumptions, often limiting flexibility and interpretability. We introduce a "model-free" framew…
- arXiv cs.AIagentsHAFixAgent: History-Aware Agentic Program Repair
arXiv:2511.01047v4 Announce Type: replace-cross Abstract: Automated program repair (APR) has recently shifted toward large language models and agent-based systems, while prior work shows that repository history can help…
- arXiv cs.AIagentsPreference-Conditioned Multi-Objective Reinforcement Learning for Tactical Decision Making in Heavy-Duty Vehicles
arXiv:2601.18783v3 Announce Type: replace-cross Abstract: A central difficulty in applying reinforcement learning to real-world applications such as autonomous driving is that conventional scalar reward formulations obta…
- arXiv cs.AIresearchCOBALT: Contextual Bandit Learning with Offline Trajectories for Multi-Turn Code Generation
arXiv:2602.03806v2 Announce Type: replace-cross Abstract: Training large language models (LLMs) for multi-turn code generation typically relies on online reinforcement learning (RL), in which models learn from repeatedly…
- arXiv cs.AIresearchBeyond State-Wise Mirror Descent: Offline Policy Optimization with Parametric Policies
arXiv:2602.23811v5 Announce Type: replace-cross Abstract: We investigate the theoretical aspects of offline reinforcement learning (RL) under general function approximation. While prior works (e.g., Xie et al., 2021) hav…