AI news archive.
4,959 articles filed under Agents, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsThe Agentic Scientist: Orchestrating Coding Agents for Automated Scientific (Co-)Discovery
arXiv:2610.18933v1 Announce Type: new Abstract: Modern coding agents are fundamentally reshaping the role of human input for computational research: they write code, launch in silico experiments, analyze their results, a…
- arXiv cs.AIagentsSmart Alone, Blind Together: Why Accuracy Fails to CertifyTheory of Mind in Multi-Agent Teams
arXiv:2610.19051v1 Announce Type: new Abstract: Coordinators that act on what they infer about teammates are usually certified by the accuracy of those inferences. We argue that accuracy is the wrong certificate, for two…
- arXiv cs.AIagentsPlan or Retrieve First? A Scale Study of Tool Discovery Architectures
arXiv:2610.19068v1 Announce Type: new Abstract: Tool-augmented LLM agents discover relevant APIs from large registries before planning execution, yet finite context windows prohibit injecting all tool definitions into a…
- arXiv cs.AIagentsGroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning
arXiv:2610.19100v1 Announce Type: new Abstract: Constructing robust GUI agents relies on multimodal visual grounding: the ability to precisely map natural language intents to actionable on-screen components. Despite abun…
- arXiv cs.AIagentsPlant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders
arXiv:2610.19131v1 Announce Type: new Abstract: Maintenance work orders are the largest written record of how industrial equipment fails, yet in most plants they are close to unusable for analysis. Short texts are abbrev…
- arXiv cs.AIagentsAgentCIScope: Grounded Contextual Integrity Evaluation for LLM Agents
arXiv:2610.19158v1 Announce Type: new Abstract: LLM agents are becoming increasingly capable of handling daily workspace tasks by interacting with application environments through tool calls. However, as agents read and…
- arXiv cs.AIagentsInto the Unknown: Benchmarking AI agents on open scientific research problems
arXiv:2610.19159v1 Announce Type: new Abstract: Can AI agents carry out research on open scientific problems? We introduce Into the Unknown, a benchmark that provides an agent with a science paper along with a research d…
- arXiv cs.AIagentsVerification Turns Capability Gaps into Routing Decisions: Small Language Models for Clinical Documentation
arXiv:2610.19169v1 Announce Type: new Abstract: Clinical note generation from doctor--patient dialogue is high-volume and hallucination-intolerant: a fabricated laboratory value can create patient-safety risk, not a qual…
- arXiv cs.AIagentsIdentity, Incentives, and Cooperative Behavior for AI: Reverse-Engineering Multiagent Cooperation
arXiv:2610.19188v1 Announce Type: new Abstract: The behavior of AI systems increasingly depends on the extent to which agents cooperate or coordinate with one another, in ways that may be desirable or not depending on th…
- arXiv cs.AIagentsAgentic Self-Awareness: Aligning Knowing, Acting, and Holding
arXiv:2610.19226v1 Announce Type: new Abstract: Frontier AI agents increasingly act in environments where seemingly solvable tasks may lack prerequisites, contain conflicting requirements, or reveal blockers only during…
- arXiv cs.AIagentsMemeTrace: A Multimodal LLM-Based Agent for Safety-Aware Meme Provenance Analysis
arXiv:2610.19267v1 Announce Type: new Abstract: Online memes are often carriers of unsafe content such as disinformation and propaganda. Their rapid evolution in response to news events and strong dependence on contextua…
- arXiv cs.AIagentsWhose turn is it? Context-guided model routing for stateful agents
arXiv:2610.19292v1 Announce Type: new Abstract: Model routing can reduce the cost of LLM applications by assigning simpler requests to cheaper models without sacrificing the quality of the response. However, most routers…
- arXiv cs.AIagentsWhen Is Smaller Greener? Right-Sizing Language Models for Agentic AI
arXiv:2610.19345v2 Announce Type: replace Abstract: AI agents use a large language model (LLM) to plan and act over many steps. Which model should an agent use, so that its tasks are completed reliably with the least ene…
- arXiv cs.AIagentsSeeing, Maintaining, and Learning: The Evolution of Multimodal Agent Memory
arXiv:2610.19366v1 Announce Type: new Abstract: Multimodal models and agents have emerged to enable AI systems to learn from and interact with the real world more like humans by integrating information across multiple mo…
- arXiv cs.AIagentsDeLoop: Repeated Actions as Self-Supervision for Long-Horizon Agents
arXiv:2610.19376v1 Announce Type: new Abstract: Language-model agents that run long tasks often redo work they have already done: they re-issue a tool call they made earlier and get back exactly the same observation, a s…
- arXiv cs.AIagentsEvoWeave: Composing Reusable Agent Experience through Operation Level Graphs
arXiv:2610.19381v1 Announce Type: new Abstract: Different tasks often require interactive agents to perform similar operations at intermediate steps. Retrieving an entire past workflow can introduce irrelevant steps, whe…
- arXiv cs.AIagentsixi-sLM: On-Device sLM for AI Call Agent Service
arXiv:2610.19382v1 Announce Type: new Abstract: Running an AI call agent on-device makes it private, network-independent, and free of per-query cloud cost, but shipping one as a real-world service imposes constraints a c…
- arXiv cs.AIagentsSTRATA: Self-Maintaining Structured Experience Graphs for Self-Evolving Language Agents
arXiv:2610.19386v1 Announce Type: new Abstract: Large language model (LLM) agents can self-evolve at test time by accumulating experience: retrieving relevant memories before acting and writing new lessons back afterward…
- arXiv cs.AIagentsEpiGuard: An Autonomous Infectious-Disease Early Warning Agent Platform
arXiv:2610.19396v1 Announce Type: new Abstract: Infectious disease outbreaks often spread rapidly before signs of widespread transmission become apparent, leaving a big challenge for rapid response and effective interven…
- arXiv cs.AIagentsSERO: Learning from Executable Oracle Continuations for Long-Horizon Agentic Reinforcement Learning
arXiv:2610.19397v1 Announce Type: new Abstract: Long-horizon agentic RL is constrained by the futures visited by the current policy. When rollouts fail to reach useful behaviors, improved credit assignment cannot provide…
- arXiv cs.AIagentsAgentSafeEvo: Adversarial Co-Evolution for LLM Agent Safety
arXiv:2610.19422v1 Announce Type: new Abstract: Large language model (LLM) agents accomplish complex tasks through multi-turn interactions with external tools. However, malicious content embedded in user queries, tool de…
- arXiv cs.AIagentsEvaluating Agentic Capabilities Before Post-Training: An End-to-End Approach
arXiv:2610.19425v1 Announce Type: new Abstract: Agentic capabilities are increasingly important for modern LLMs, yet remain difficult to evaluate during pretraining. Existing approaches typically rely on static trajector…
- arXiv cs.AIagentsEvoProbe: Evaluation-Driven Benchmark Evolution for LLM Agents
arXiv:2610.19433v1 Announce Type: new Abstract: Static benchmarks for large language model (LLM) agents face saturation and data contamination risks, motivating continual updates to evaluation tasks. However, aggregate s…
- arXiv cs.AIagentsAdCoEvo: A Co-Evolving Multi-Agent Advertising Delivery System
arXiv:2610.19454v1 Announce Type: new Abstract: Digital advertising is an important means for platforms to reach potential users and drive business growth. Advertising delivery is the core link between marketing objectiv…
- arXiv cs.AIagentsPFedMA: Who Should Join Next as Federated Multi-Agent Orchestration Evolves?
arXiv:2610.19458v1 Announce Type: new Abstract: Federated Multi-Agent Systems (FedMAS) offer a promising paradigm for collaborative reasoning among LLM-based agents while keeping raw client data decentralized. However, r…
- arXiv cs.AIagentsMemCode: Discrete Semantic Representations for Long-Term Agent Memory
arXiv:2610.19472v1 Announce Type: new Abstract: Long-term memory remains a fundamental bottleneck for large language model (LLM) agents across sessions. Existing systems typically adopt a write-then-retrieve paradigm, st…
- arXiv cs.AIagentsWhen Agent Memory Needs to Compute: Evidence Access and Executable Aggregation
arXiv:2610.19474v2 Announce Type: replace Abstract: Questions about an agent's history often require aggregating events, which retrieval alone does not resolve. We combine source-preserving evidence access with executabl…
- arXiv cs.AIagentsStructure Before Intelligence: Gating What Agentic Scientific Workflows Release
arXiv:2610.19507v1 Announce Type: new Abstract: An agentic scientific workflow can run to completion and still release wrong results: a silent numerical error, a stale input, or a defect in a shared upstream step can con…
- arXiv cs.AIagentsLabCompiler: A Framework for Verifiable Human-Machine Laboratory Execution from Imperfect Protocols
arXiv:2610.19550v1 Announce Type: new Abstract: Closing the loop of AI-driven science requires reliably translating natural-language protocols into executable laboratory workflows despite protocol defects and heterogeneo…
- arXiv cs.AIagents$\tau$-Emotion-Bench: Exposing the Interaction Fragility of LLM Agents through Real-World User Emotions
arXiv:2610.19560v1 Announce Type: new Abstract: In long-horizon tool-calling tasks, LLM agents can fail not only because of tool-use errors, but also because of ineffective interaction with users. Yet existing benchmarks…
- arXiv cs.AIagentsJockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making
arXiv:2610.19589v1 Announce Type: new Abstract: Repeated event forecasting requires extracting predictive signals from noisy evidence and evaluating their value across successive decisions. We introduce JockeyBench, a be…
- arXiv cs.AIagentsL3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents
arXiv:2610.19591v1 Announce Type: new Abstract: The lasting gains in AI have come from methods that absorb more computation, and language models already scale computation within a call. Useful work, though, happens acros…
- arXiv cs.AIagentsModel What Matters: Decision-Guided Selective World Modeling for Language Agents
arXiv:2610.19642v1 Announce Type: new Abstract: Reinforcement learning provides a direct mechanism for optimizing task objectives in large language model-driven agents, but task rewards are typically sparse trajectory-le…
- arXiv cs.AIagentsMCAIR: Monitoring and Correcting Agent Unfaithfulness via Internal Representations
arXiv:2610.19679v1 Announce Type: new Abstract: Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in completing complex tasks through multi-step reasoning and tool interactions. However, r…
- arXiv cs.AIagentsC3: Rethinking Process Supervision via Confidence-Guided Credit Correction in Multi-Turn Reinforcement Learning
arXiv:2610.19695v1 Announce Type: new Abstract: Large language models are increasingly trained as multi-turn agents that interact with external tools. However, outcome-only reinforcement learning can produce a form of mi…
- arXiv cs.AIagentsRuleTrek: Benchmarking Agentic Rule Reasoning in Complex Scenarios
arXiv:2610.19698v1 Announce Type: new Abstract: Large language models (LLMs) increasingly tackle complex real-world tasks in specialized domains. However, identifying and correctly applying domain-specific rules remains…
- arXiv cs.AIagentsLearning Perception to Action Rules through Reinforcement Learning for Theory-of-Mind Reasoning
arXiv:2610.19796v1 Announce Type: new Abstract: In Theory-of-Mind (ToM) reasoning, large language models may correctly infer the agent's mental state but fail to use that inference when predicting or judging the agent's…
- arXiv cs.AIagentsReproBAIT: Can coding agents build their own biological AI tools?
arXiv:2610.19810v1 Announce Type: new Abstract: Large language models are increasingly competent at leveraging scientific knowledge and existing computational tools to execute complex workflows. At present, little inform…
- arXiv cs.AIagentsOne Monitor Does Not Fit All: Formalism-Agnostic Runtime Verification for Agentic AI
arXiv:2610.19818v1 Announce Type: new Abstract: Agentic AI systems combine model decisions, mutable state, tool calls, and delegation to external agents. Monitoring them requires observations from different execution bou…
- arXiv cs.AIagentsActionOPD: Distilling Decisions, Not Tokens
arXiv:2610.19820v1 Announce Type: new Abstract: Distilling capable language-model agents into smaller models requires aligning the student's behavior with the teacher's. This raises a fundamental question: at which state…
- arXiv cs.AIagentsTeaching Agents in their Own Style: Blackbox Teacher Distillation for Web Agents
arXiv:2610.19834v1 Announce Type: new Abstract: Frontier web agents using closed models have high inference and computational costs. Smaller open models can be an alternative, but to be performant, they require training…
- arXiv cs.AIagentsLAGO: Language Agents with Reusable Growing Options
arXiv:2610.19843v1 Announce Type: new Abstract: Long-horizon tasks expose a weakness of monolithic tool-use loops: one agent must decompose, execute, and verify a task within a single trajectory. We introduce LAGO, which…
- arXiv cs.AIagentsEuler: Towards Human-Centered Multi-Agent Mathematical Discovery
arXiv:2610.19864v1 Announce Type: new Abstract: Mathematical research drives scientific discovery, and recent advances in AI for mathematics have shown that language-model agents can solve challenging problems and develo…
- arXiv cs.AIagentsExplorer-Zero: Improving Self-Play for Long-Horizon Tasks with Decomposable Verifiable Rewards
arXiv:2610.19868v1 Announce Type: new Abstract: Training LLM agents on long-horizon interactive tasks with reinforcement learning typically relies on human-annotated tasks and reliable verification, both expensive to con…
- arXiv cs.AIagentsHow GUI Agents Fail: GraceBench for Execution Stability Diagnosis
arXiv:2610.19900v1 Announce Type: new Abstract: As GUI agents are increasingly deployed for mobile task automation, their reliability depends not only on final task outcomes, but also on whether they reach these outcomes…
- arXiv cs.AIagentsNeurosymbolic Learning of Interpretable World Models
arXiv:2610.19955v1 Announce Type: new Abstract: World models aim to learn compact representations of an environment from observations, enabling agents to predict future states and plan their behaviour. However, most exis…
- arXiv cs.AIagentsWhat Coding Agents Leave Unsaid: Measuring Task-State Gaps Between Execution and Reports
arXiv:2610.20003v2 Announce Type: replace Abstract: Human supervisors rely on coding agents' reports to assess progress, verify repairs, and decide when to intervene. But a supervisor cannot investigate a failure they we…
- arXiv cs.AIagentsMemorizing Who, Knowing How: Benchmarking Personal and Procedural Memory for Persistent Assistants
arXiv:2610.20022v2 Announce Type: replace Abstract: Memory systems for persistent personal agents must move beyond isolated memories, maintaining both personal memory about whom they serve and procedural memory about how…
- arXiv cs.AIagentsEvalBuilder: Can LLMs Build Efficient Benchmarks?
arXiv:2610.20099v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly used for long-horizon research and engineering, yet evaluating these capabilities remains expensive, time-consuming, and…
- arXiv cs.AIagentsMaximising Transparency in Context Compaction Improves Retention of Facts and Commitments for Lightweight Coding Agents
arXiv:2610.20116v1 Announce Type: new Abstract: Long coding agent sessions outgrow the model's context window, so agents compact: older history is replaced by something shorter, usually a model-written summary. A summary…