AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsSkillCommit: Evolving Agent Skills through Behaviorally Validated Scope Expansion
arXiv:2608.15165v2 Announce Type: replace Abstract: Large language model (LLM) agents can continually improve without parameter updates by converting historical experience into reusable procedural knowledge. However, exi…
- arXiv cs.AIresearchSPAR-Hate: Auditor-Guided Multi-Perspective Role Reasoning for Bilingual Hate Speech Parsing
arXiv:2608.22018v4 Announce Type: replace Abstract: Hate speech research has moved from coarse-grained classification towards structured parsing, where systems jointly identify targets, supporting arguments, and target-l…
- arXiv cs.AIagentsCDEG: Learning Decision-Critical Evidence for Long-Horizon Diagnostic Agents
arXiv:2608.22899v3 Announce Type: replace Abstract: Unlike static medical question answering, long-horizon diagnosis captures the sequential nature of clinical practice: evidence is progressively acquired, integrated, an…
- arXiv cs.AIresearchRight Values, Wrong Places: Diagnosing and Fixing Record Misplacement in LLM-Generated JSON
arXiv:2608.25358v2 Announce Type: replace Abstract: Large language models can generate valid JSON that preserves field values yet binds them to the wrong records or hierarchy levels. We introduce JSONPlacement, a control…
- arXiv cs.AIresearchFaithSieve: Fine-Grained Evaluation of Math Proofs with Faithful Formal Evidence
arXiv:2608.26310v2 Announce Type: replace Abstract: Large language models can now generate complex, multi-step mathematical proofs, but reliably determining their correctness and localizing early logical errors remains a…
- arXiv cs.AIresearchLLMs Can Design Near-Optimal OR Algorithms
arXiv:2608.27296v3 Announce Type: replace Abstract: We ask whether large language models (LLMs) can design effective algorithms for well-specified operations research (OR) problems. We study inventory control, queueing n…
- arXiv cs.AIresearchSHAPE of Thinking: Mathematical Heuristics for Analyzing and Improving LLM Reasoning
arXiv:2608.28600v2 Announce Type: replace Abstract: Large language models (LLMs) achieve strong performance on mathematical reasoning benchmarks, yet final-answer accuracy reveals little about the strategies they use. We…
- arXiv cs.AIresearchSame Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts
arXiv:2609.00578v2 Announce Type: replace Abstract: Cybersecurity assistance is dual use, requiring decisions about when to answer, limit, or refuse a request. Prior work studies refusal evaluation and conversational jai…
- arXiv cs.AIagentsOmniStepBench: Deployment-Aware Agent Evaluation across Modalities and Task Complexity
arXiv:2609.06059v2 Announce Type: replace Abstract: As large language models evolve from question-answering systems into general-purpose agents, evaluation must move beyond static answer correctness to assess multimodal…
- arXiv cs.AIresearchWhen Do Language Models Add Value? Relative Competence in Hybrid Event Forecasting
arXiv:2609.12101v2 Announce Type: replace Abstract: When a forecasting system already has a prediction from a market, crowd, or statistical model, when should it use a language model? Standalone accuracy cannot answer th…
- arXiv cs.AIresearchBeyond Size: How Gradients Shape Pruning Decisions in Large Language Models
arXiv:2311.04902v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) with billions of parameters are prime targets for network pruning, which removes some model weights without hurting performance. Prio…
- arXiv cs.AIresearchForget About It: Mitigating Forgetting Without Replay or Auxiliary Objectives
arXiv:2501.05559v2 Announce Type: replace-cross Abstract: Have we made mitigating catastrophic forgetting more complicated than it needs to be? Continual adaptation often degrades performance on earlier tasks, motivating…
- arXiv cs.AIresearchVideo Depth Anything: Consistent Depth Estimation for Super-Long Videos
arXiv:2501.12375v4 Announce Type: replace-cross Abstract: Depth Anything has achieved remarkable success in monocular depth estimation with strong generalization ability. However, it suffers from temporal inconsistency i…
- arXiv cs.AIresearchTrustworthy AI Must Account for Interactions
arXiv:2504.07170v3 Announce Type: replace-cross Abstract: Trustworthy AI encompasses many aspirational aspects for aligning AI systems with human values, including fairness, privacy, robustness, explainability, and uncer…
- arXiv cs.AIresearchSynergistic Vision-Language Reinforcement Enables Scalable On-Demand Analysis across Diverse Clinical Tasks
arXiv:2505.03380v3 Announce Type: replace-cross Abstract: Accurate segmentation of tumors and organs-at-risk is essential for oncology but remains time-consuming and expertise-intensive. Existing AI systems often require…
- arXiv cs.AIresearchRule-based and reinforcement-learning policies for resource allocation in a pediatric echo lab: a simulation study
arXiv:2506.06297v2 Announce Type: replace-cross Abstract: Scheduling echocardiographic examinations in a hospital setting presents significant challenges due to non-deterministic factors such as patient no-shows, variabl…
- arXiv cs.AIresearchIs This Just Fantasy? Language Model Representations Reflect Human Judgments of Event Plausibility
arXiv:2507.12553v4 Announce Type: replace-cross Abstract: Language models (LMs) are used for a diverse range of tasks, from question answering to writing fantastical stories. In order to reliably accomplish these tasks,…
- arXiv cs.AIresearchThe Invisible Leash: Why RLVR May or May Not Escape Its Origin
arXiv:2507.14843v5 Announce Type: replace-cross Abstract: Recent advances highlight Reinforcement Learning with Verifiable Rewards (RLVR) as a promising approach to enhancing LLM capabilities. However, it remains unclear…
- arXiv cs.AIresearchAlign Where the Words Look: Cross-Attention-Guided Patch Alignment with Contrastive and Transport Regularization for Bengali Captioning
arXiv:2509.18369v2 Announce Type: replace-cross Abstract: Adapting vision-language models in low-resource languages remains challenging, as they often produce fluent text about the wrong objects. This stems from scarce p…
- arXiv cs.AIresearchAnticipatory Evaluation of Language Models: Predicting LLM Benchmark Scores from Descriptions
arXiv:2509.20645v4 Announce Type: replace-cross Abstract: Modern ML research increasingly faces an evaluation bottleneck: building benchmarks and running experiments is costly, slowing iteration. We ask how much can be i…
- arXiv cs.AIagentsEigen-Agent: Adaptive Multi-Agent Scientific Reasoning with Monitor-Based RAG
arXiv:2509.21193v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have recently shown strong progress on scientific reasoning, yet two major bottlenecks remain. First, explicit retrieval fragments re…
- arXiv cs.AIagentsIt's My Agent, Not Me: Personal Privacy Boundaries in AI-Delegated Information Sharing
arXiv:2509.21712v2 Announce Type: replace-cross Abstract: Aligning AI systems with human privacy preferences requires understanding individuals' nuanced disclosure behaviors beyond general privacy norms. Yet eliciting th…
- arXiv cs.AIresearchContext Value Informed In-Context Reinforcement Learning
arXiv:2509.23209v2 Announce Type: replace-cross Abstract: In-context reinforcement learning (ICRL) has emerged as an effective paradigm for test time adaptation to unseen tasks without parameter updates. However, existin…
- arXiv cs.AIagentsGeneralized Multi-agent Social Simulation Framework
arXiv:2510.06225v2 Announce Type: replace-cross Abstract: Multi-agent social interaction has clearly benefited from Large Language Models. However, current simulation systems still face challenges such as difficulties in…
- arXiv cs.AIresearchEconCausal: A Context-Aware Economic Reasoning Benchmark for Large Language Models
arXiv:2510.07231v5 Announce Type: replace-cross Abstract: Socio-economic causal effects depend heavily on their institutional and environmental contexts. The same intervention can produce different, even opposite, effect…
- arXiv cs.AIresearchLLMs Can Get "Brain Rot": A Pilot Study on Twitter/X
arXiv:2510.13928v3 Announce Type: replace-cross Abstract: We propose and test the LLM Brain Rot Hypothesis: continual exposure to junk web text induces lasting cognitive decline in large language models (LLMs). To unveil…
- arXiv cs.AIresearchChain of Time: In-Context Physical Simulation with Image Generation Models
arXiv:2511.00110v2 Announce Type: replace-cross Abstract: We propose a novel cognitively-inspired prompting method to stabilize and interpret the physical simulation produced by autoregressive image generation models, ca…
- arXiv cs.AIresearchWhen is a System Discoverable from Data? Discovery Requires Chaos
arXiv:2511.08860v3 Announce Type: replace-cross Abstract: The deep learning revolution has enabled major AI advances in physical sciences, particularly in discovering dynamical systems from observational data. Yet the re…
- arXiv cs.AIresearchVEDA: Generation of 3D Molecules via Variance-Exploding Diffusion with Annealing
arXiv:2511.09568v3 Announce Type: replace-cross Abstract: Diffusion models show promise for 3D molecular generation, but face a fundamental trade-off between sampling efficiency and conformational accuracy. While flow-ba…
- arXiv cs.AIresearchWhat If They Took the Shot? A Hierarchical Bayesian Framework for Counterfactual Expected Goals
arXiv:2511.23072v2 Announce Type: replace-cross Abstract: How would the probability of scoring change if a different player took the same shot? Although expected goals models describe the difficulty of scoring opportunit…
- arXiv cs.AIresearchThe Instability of Safety: How Random Seeds and Temperature Expose Inconsistent LLM Refusal Behavior
arXiv:2512.12066v4 Announce Type: replace-cross Abstract: Current safety evaluations of large language models rely on single-shot testing, implicitly assuming that model responses are deterministic and representative of…
- arXiv cs.AIresearchSchr\"odinger's Navigator: Imagining an Ensemble of Futures for Zero-Shot Object Navigation
arXiv:2512.21201v5 Announce Type: replace-cross Abstract: Zero-shot object navigation (ZSON) requires robots to find target objects in unseen environments without task-specific fine-tuning or pre-built maps, a key capabi…
- arXiv cs.AIresearchImaging-anchored multiomics in cardiovascular disease: integrating cardiac imaging, bulk, single-cell, and spatial transcriptomics
arXiv:2601.07871v3 Announce Type: replace-cross Abstract: Cardiovascular disease arises from interactions between inherited risk, molecular programmes, and tissue-scale remodelling that are observed clinically through im…
- arXiv cs.AIagentsSWE-Spot: Building Small Repo-Experts with Repository-Centric Learning
arXiv:2601.21649v2 Announce Type: replace-cross Abstract: The deployment of coding agents in privacy-sensitive and resource-constrained environments drives the demand for capable open-weight Small Language Models (SLMs).…
- arXiv cs.AIresearchDeep Time-series Forecasting Needs Kernelized Moment Balancing
arXiv:2602.00717v2 Announce Type: replace-cross Abstract: Deep time-series forecasting can be formulated as a distribution balancing problem aimed at aligning the distribution of the forecasts and ground truths. Accordin…
- arXiv cs.AIresearchParameter-Efficient Temporal Domain Generalization
arXiv:2602.11965v2 Announce Type: replace-cross Abstract: Temporal domain generalization (TDG) learns from a sequence of observed domains to predict models for unseen future domains. Existing parameter-dynamics methods b…
- arXiv cs.AIagentsNePPO: Near-Potential Policy Optimization for General-Sum Multi-Agent Reinforcement Learning
arXiv:2603.06977v3 Announce Type: replace-cross Abstract: Multi-agent reinforcement learning (MARL) is increasingly used to design learning-enabled agents that interact in shared environments. However, training MARL algo…
- arXiv cs.AIresearchLLMON: An LLM-native Markup Language to Leverage Structure and Semantics at the LLM Interface
arXiv:2603.22519v4 Announce Type: replace-cross Abstract: Textual Large Language Models (LLMs) provide a simple and familiar interface: a string of text is used for both input and output. However, the information conveye…
- arXiv cs.AIresearchPLACID: Privacy-preserving Large language models for Acronym Clinical Inference and Disambiguation
arXiv:2603.23678v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) offer transformative solutions across many domains, but healthcare integration is hindered by strict data privacy constraints. Clinic…
- arXiv cs.AIresearchA Wireless World Model for AI-Native 6G Networks
arXiv:2603.25216v2 Announce Type: replace-cross Abstract: Integrating AI into the physical layer is a cornerstone of 6G networks. However, current data-driven approaches struggle to generalize across dynamic environments…
- arXiv cs.AIresearchCan We Change "Stroke Size" to Simplify Regression Learning?
arXiv:2603.26783v5 Announce Type: replace-cross Abstract: Deep-learning-based image generation is increasingly useful but expensive. Across the different timesteps in the denoising chain, high-noise timesteps present dis…
- arXiv cs.AIresearchMapping data literacy trajectories in K-12 education
arXiv:2603.28317v2 Announce Type: replace-cross Abstract: Data literacy skills are fundamental in computer science education. However, understanding how data-driven systems work represents a paradigm shift from tradition…
- arXiv cs.AIagentsTerminal Agents Suffice for Enterprise Automation
arXiv:2604.00073v4 Announce Type: replace-cross Abstract: There has been growing interest in building agents that can interact with digital platforms to execute meaningful enterprise tasks autonomously. Among the approac…
- arXiv cs.AIresearchExplorable Proofs: Making Written Proofs Explorable by Grounding Them in Formal Representations
arXiv:2604.02598v3 Announce Type: replace-cross Abstract: LLM-generated explanations can make technical content more accessible, but there is a ceiling on what they can support interactively. Because LLM outputs are stat…
- arXiv cs.AIagentsTowards Secure Agent Skills: Architecture, Threat Taxonomy, and Security Analysis
arXiv:2604.02837v2 Announce Type: replace-cross Abstract: Agent Skills is an emerging open standard that defines a modular, filesystem-based packaging format enabling LLM-based agents to acquire domain-specific expertise…
- arXiv cs.AIresearchAdaptFuse: Training-Free Sequential Preference Inference via Externalized Bayesian Computation
arXiv:2604.03925v2 Announce Type: replace-cross Abstract: Large language models (LLMs) struggle to accumulate evidence across multiple rounds of user interactions, failing to update their beliefs in a manner consistent w…
- arXiv cs.AIresearchSnapFlow: One-Step Action Generation for Flow-Matching VLAs via Progressive Self-Distillation
arXiv:2604.05656v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models with flow-matching action heads, such as {\pi}0 and {\pi}0.5, have become a standard recipe for generalist robot manipulation,…
- arXiv cs.AIagentsAutomated Standardization of Legacy Biomedical Metadata Using an Ontology-Constrained LLM Agent
arXiv:2604.08552v5 Announce Type: replace-cross Abstract: Descriptive scientific metadata in public repositories are often incomplete and inconsistent with community standards and ontologies, limiting data FAIRness. Larg…
- arXiv cs.AIagentsLatent Preference Modeling for Multi-Session Personalized Tool Calling
arXiv:2604.17886v3 Announce Type: replace-cross Abstract: Users often omit essential details in their requests to LLM-based agents, resulting in under-specified inputs for tool use. This poses a fundamental challenge for…
- arXiv cs.AIresearchDocument-as-Image Representations Fall Short for Scientific Retrieval
arXiv:2604.18508v2 Announce Type: replace-cross Abstract: Many recent document embedding models are trained on document-as-image representations, embedding rendered pages as images rather than the underlying source. Mean…