AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsWhere Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security
arXiv:2610.07657v2 Announce Type: replace Abstract: LLM-based multi-agent systems (MAS) engage tools, share memory, and delegate tasks, often encountering adversarial content. Current defenses for MAS are typically evalu…
- arXiv cs.AIresearchMassive Activation Gating Channel in Large Language Models
arXiv:2610.07661v1 Announce Type: new Abstract: Massive activations, a phenomenon in which a small number of hidden channels exhibit exceptionally large magnitudes, are pervasive in large language models (LLMs). However,…
- arXiv cs.AIagentsEIO-Agents: The Missing Semantic Layer for AI Agent Evaluation
arXiv:2610.07675v1 Announce Type: new Abstract: AI agents are entering production in increasingly consequential environments without a shared semantic standard for what their evaluations actually mean. Scores, traces, ju…
- arXiv cs.AIresearchBluffJAX: Adversarial Imperfect Information Games in JAX
arXiv:2610.07686v1 Announce Type: new Abstract: We introduce BluffJAX: an open-source suite of adversarial imperfect information games in JAX. We provide canonical implementations of games designed for high simulation th…
- arXiv cs.AIresearchOn the Boundary of Admission Gates: An Injected-Truth Study of Falsification-First Selection in Quantitative Strategy Research
arXiv:2610.07701v1 Announce Type: new Abstract: Strategy research conflates two problems: finding a profitable rule, and establishing that the finding is not search luck. The latter calls for admission gates -- statistic…
- arXiv cs.AIagentsAgentMemGate: Addressing Speculation Contamination in Conversational Assistant Memory
arXiv:2610.07707v1 Announce Type: new Abstract: Conversational AI assistants with long-term memory extract facts from user messages into a store consulted in later conversations. A stated plan can enter that store as fac…
- arXiv cs.AIresearchEvidence Before Sampling: Interpretable Implicit Negative Candidate Discovery for Recommendation
arXiv:2610.07708v2 Announce Type: replace Abstract: Recommender systems learn from observed user-item interactions, but explicit negative feedback is often unavailable. Since deep learning models require negative signals…
- arXiv cs.AIagentsPERSIST: Who-What-When Memory Across Sessions for Full-Duplex Spoken Dialogue
arXiv:2610.07725v1 Announce Type: new Abstract: Modern voice assistants may be shared by multiple users and should be able to answer questions about earlier conversations such as "When did I originally plan to leave?" or…
- arXiv cs.AIagentsHow Well Do LLMs Reason with Noisy Evidence? An Active Visual Reasoning Benchmark
arXiv:2610.07751v1 Announce Type: new Abstract: Real-world reasoning rarely reduces to static question answering: agents must actively gather information from tools and sensors that are often noisy and unreliable. Yet mo…
- arXiv cs.AIagentsST-Bench: A Spatial-Temporal Benchmark for Multi-Agent System Generation on Scientific Research Tasks
arXiv:2610.07763v1 Announce Type: new Abstract: The rapid progress of LLM-based multi-agent systems (MAS) has shown that they largely outperform single agents on coding, math, and QA tasks, where executable tests provide…
- arXiv cs.AIresearchOTel: Open Telco AI Datasets, Benchmarks, and Models
arXiv:2610.07766v5 Announce Type: replace Abstract: We present Open Telco (OTel), an open telecom AI resource that releases derived telecom datasets for retrieval, reranking, instruction tuning, and safety/abstention, to…
- arXiv cs.AIagentsQuantization Effects on Tool-Failure Recovery Vary Across Prompts and Evaluation Designs
arXiv:2610.07781v1 Announce Type: new Abstract: Post-training quantization reduces the cost of deploying language-model agents, but its effect on recovery from temporary tool failures can depend on how recovery is evalua…
- arXiv cs.AIagentsPersistent Memory in Multi-Agent LLM Inference: What It Costs, What It Buys, and When You Can Tell
arXiv:2610.07782v1 Announce Type: new Abstract: Decomposing long-context inference across cooperating agents bounds the active KV cache per call rather than total evidence, which matters when KV-cache memory binds. Many…
- arXiv cs.AIagentsAttacca: Goal-Directed Control under State Continuity for Long-Horizon Embodied Agents
arXiv:2610.07785v1 Announce Type: new Abstract: A central capability of embodied agents is to accomplish complex objectives through sequences of interdependent tasks. Yet existing visual goal-conditioned policies underly…
- arXiv cs.AIagentsOOPMAS: Object-Oriented Multi-Agent Systems for Query-Level Workflow Generation
arXiv:2610.07787v1 Announce Type: new Abstract: Multi-agent systems (MAS) powered by large language models have shown strong performance across code generation, mathematical reasoning, and question answering. However, ex…
- arXiv cs.AIresearchIllusory Pattern Perception Drives Spurious Inference in Large Language Models
arXiv:2610.07791v1 Announce Type: new Abstract: Illusory pattern perception is a well-documented human cognitive tendency to infer meaningful relationships in data that is actually random. Such a tendency, often describe…
- arXiv cs.AIresearchThin Evidence, Thick Priors: How Language Models Substitute Identity for Missing Financial Facts
arXiv:2610.07798v1 Announce Type: new Abstract: People increasingly ask large language models what to do with their money, yet seldom describe their finances in full. This paper asks what a model does with the gap. Holdi…
- arXiv cs.AIresearchThinkFuse: Trajectory-Aware Test-Time Fusion for Small Reasoning Models
arXiv:2610.07803v1 Announce Type: new Abstract: Small reasoning models (SRMs) have shown strong performance on complex reasoning tasks by generating extended chain-of-thought trajectories, but they often fail to recover…
- arXiv cs.AIagentsDo I Need the Cloud? Uncertainty-Aware Step-Level Handoff for Small Language Model Agents
arXiv:2610.07816v1 Announce Type: new Abstract: Small language models (SLMs) are attractive as local agent controllers because they reduce remote inference, latency, and deployment footprint, yet structured tool errors c…
- arXiv cs.AIagentsAgentic Semantic Sensing for Resource-Adaptive AI-RAN
arXiv:2610.07829v1 Announce Type: new Abstract: Semantic sensing (SemS) acquires task-relevant information rather than reconstructing complete physical information. Existing SemS formulations typically operate open loop:…
- arXiv cs.AIagentsDHCG: Dynamic Construction of Hierarchical Collaboration Graphs for LLM-Based Multi-Agent Reasoning
arXiv:2610.07835v1 Announce Type: new Abstract: LLM-based multi-agent systems (MAS) have demonstrated strong capabilities in solving complex problems across diverse domains. Recently, the dynamic orchestration of agent s…
- arXiv cs.AIagentsRA-MoWE: Workflow-Affinity Embeddings for Query Clustering and Agentic Workflow Generation
arXiv:2610.07851v1 Announce Type: new Abstract: Agentic workflows enable large language models (LLMs) to solve complex tasks by coordinating reasoning, tool use, and verification. However, a workflow optimized for an ent…
- arXiv cs.AIagentsWorkflowOps: Learning Agent Collaboration Priors for Multi-Agent Workflow Orchestration
arXiv:2610.07860v1 Announce Type: new Abstract: Multi-agent systems are increasingly deployed for complex knowledge work, yet their orchestration layers remain largely memoryless: each new task is decomposed, assigned, a…
- arXiv cs.AIagentsSelf-Referenced Social Preferences: Cooperation without Observing Others Rewards
arXiv:2610.07881v2 Announce Type: replace Abstract: Social preferences can promote cooperation in multi-agent reinforcement learning, but existing approaches often require agents to observe the rewards of their peers. In…
- arXiv cs.AIagentsShanLiangRen: A Nutrition Agent for Personalized Daily Meal Planning
arXiv:2610.07886v1 Announce Type: new Abstract: Dietary nutrition planning plays an important role in chronic disease management and maintaining a healthy body. In applications, it must simultaneously satisfy personalize…
- arXiv cs.AIresearchTextual Environmental Context and Spatial Graphs for LLM-Based Regional SST Forecasting
arXiv:2610.07895v1 Announce Type: new Abstract: Sea surface temperature (SST) forecasting depends on local temporal persistence, regional spatial dependence, and environmental conditions that evolve with the forecast dat…
- arXiv cs.AIresearchIsotropic Yet Undecodable: The Sequential Content-Sufficiency Gap in Latent-Predictive Text Representations
arXiv:2610.07906v1 Announce Type: new Abstract: We study sequential content sufficiency by investigating whether a representation retains the ordered target information available in its input. An information-theoretic de…
- arXiv cs.AIresearchContinuous Memory Machines
arXiv:2610.07907v1 Announce Type: new Abstract: Recurrent neural networks typically compress information into a single vector-valued recurrent state, forcing short-term computation and long-term retention to share the sa…
- arXiv cs.AIagentsSIGMA: Self-Improving Alignment Generalization from a Model Spec
arXiv:2610.07935v1 Announce Type: new Abstract: LLM agents are increasingly capable of executing complex tasks and of recursively improving themselves on easy-to-verify objectives such as software engineering and mathema…
- arXiv cs.AIagentsConfidence Reasoning Graphs: Structured Confidence Estimation for LLM Agents
arXiv:2610.07948v1 Announce Type: new Abstract: When using an LLM agent in a consequential domain, making an informed decision about whether to trust its output or intervene requires calibrated confidence in the agent's…
- arXiv cs.AIagentsCan Agents Work for Everyone? Cross-User Reliability for Mobile GUI Agents in Personalized User Interfaces
arXiv:2610.07972v1 Announce Type: new Abstract: Mobile GUI agents increasingly operate on interfaces influenced by users' histories and preferences, but their reliability across different users remains underexplored. We…
- arXiv cs.AIagentsLearning from Revision Consequences: Hindsight Meta-Experience Distillation for Self-Improving Agents
arXiv:2610.07979v1 Announce Type: new Abstract: As agents continuously improve by generating and revising Skills, the process that discovers and refines those Skills becomes a learnable object in its own right. Task-Skil…
- arXiv cs.AIresearchLearning in Dreams, Winning in Reality: A Continuous Dyna Loop for a Ten-Hero MOBA
arXiv:2610.08033v1 Announce Type: new Abstract: World models are usually judged from the inside: by prediction loss, by the return a policy earns in imagination, or by how convincing their frames look. We judge one from…
- arXiv cs.AIagentsSame Feedback, Different Answer: Measuring Run-to-Run Instability in Frontier-Model Customer Feedback Analysis
arXiv:2610.08036v1 Announce Type: new Abstract: AI agents are increasingly being programmed to automate knowledge work over large collections of unstructured data. Such automation requires repeatability: when the underly…
- arXiv cs.AIagentsDAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks
arXiv:2610.08048v1 Announce Type: new Abstract: LLM agents often lack the operational knowledge to act reliably in new environments, as they must discover specific tool behaviors or environment conventions on their own.…
- arXiv cs.AIagentsSpeedrunBench: Challenging LLM Agents with Video Game Speedrunning
arXiv:2610.08076v1 Announce Type: new Abstract: Frontier LLM agents have been shown to be capable of solving increasingly complex tasks for which humans have measurable solutions. This begs the pertinent question of whet…
- arXiv cs.AIagentsSelf-Retrospection Distillation: Turning Post-hoc Experiences into Prior Foresight
arXiv:2610.08077v3 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) turns agent experience into learning signals primarily through scalar outcome rewards after interaction. For group…
- arXiv cs.AIagentsPOLAR: Ontology-Guided Risk Prevention for Tool-Calling LLM Agents
arXiv:2610.08082v1 Announce Type: new Abstract: LLM tool-use agents operate in dynamic environments where many actions carry operational risk. However, most safety mechanisms react only after errors manifest. Existing pr…
- arXiv cs.AIresearchNatural Language Questions as an Interface for Knowledge Graphs: QRAKEN Graph Distillation and Semantic Self-Healing
arXiv:2610.08095v1 Announce Type: new Abstract: Natural-language access to RDF knowledge graphs is a core Semantic Web ambition. Large language models (LLMs) have advanced Text-to-SPARQL, yet on unfamiliar graphs they of…
- arXiv cs.AIagentsBeyond Corrected Memory: Execution Consistency in Multi-Agent Systems
arXiv:2610.08101v1 Announce Type: new Abstract: Shared memory coordinates agents' actions, but correct records do not establish that those actions satisfy task requirements. Memory governance and failure diagnosis regula…
- arXiv cs.AIagentsDSV-Mem: Evaluating Multimodal Memory in Professional Workflows for MLLM Agents
arXiv:2610.08102v1 Announce Type: new Abstract: Conversational MLLM agents are increasingly expected to assist in professional workflows, from AI research and engineering design to product management and business operati…
- arXiv cs.AIagentsChartBmkAgent: Harness-Governed Multi-Agent Construction of Chart QA Benchmarks from Sparse Error-Taxonomy Specifications
arXiv:2610.08106v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) advance rapidly, while conventional benchmark development lags behind, delaying investigation of newly observed capability gaps. Su…
- arXiv cs.AIagentsTest-Time Agent Evolution for Long-Horizon Legal Reasoning
arXiv:2610.08138v1 Announce Type: new Abstract: Legal intelligence aims to support reliable decision-making across long-horizon legal processes involving evolving case states and multiple roles. However, real-world legal…
- arXiv cs.AIresearchPartially Observable Zero-shot coordination by Predicting Intention of Partner
arXiv:2610.08142v1 Announce Type: new Abstract: Zero-shot coordination in embodied settings requires acting while the partner is intermittently out of view, leaving existing methods with ambiguous partner representations…
- arXiv cs.AIresearchWhich alloy composition,what process parameters? Inferring the recipe from optimized metallic microstructure and texture
arXiv:2610.08165v1 Announce Type: new Abstract: The mechanical properties of a metallic alloy are set by its microstructure and texture: the size and shape of its grains and the orientation of their crystals. That struct…
- arXiv cs.AIresearchLFHE: Local-First Heuristic Evolution for Bounded Local Topology Search in Decentralized Learning with Non-IID Data
arXiv:2610.08176v1 Announce Type: new Abstract: Decentralized learning is highly sensitive to communication topology under non-IID data. Adaptive peer-selection methods can exploit local model information, but broader pe…
- arXiv cs.AIagentsMathematical Proof Assistants for Teaching Logic: The LogiKEy Methodology
arXiv:2610.08214v1 Announce Type: new Abstract: We report on an approach to teaching logic to mixed groups of computer science, mathematics, and philosophy students, based on the logico-pluralistic LogiKEy methodology, u…
- arXiv cs.AIagentsLearn2Play Bench: How Well Do LLM Agents Learn from Experience in Unfamiliar Environments?
arXiv:2610.08215v2 Announce Type: replace Abstract: Learning from experience is essential for LLM agents to adapt to unfamiliar and dynmaic environments. Evaluating this ability is therefore important for understanding h…
- arXiv cs.AIresearchQuantum Entangled Multimodal Fusion Networks (QEMFN): Resource-Aware Hybrid Vision-Language Fusion via Trainable Entanglement
arXiv:2610.08216v1 Announce Type: new Abstract: Multimodal vision-language systems typically fuse image and text embeddings through classical operators such as concatenation, attention, bilinear pooling, or tensor intera…
- arXiv cs.AIresearchConfidence-Ordering Reversal under Contextual Priors in Neural Decoding
arXiv:2610.08229v1 Announce Type: new Abstract: Contextual priors improve neural-to-language decoding by reshaping candidate scores. However, confidence is read from the same reshaped scores, so the errors a prior leaves…