Search: “agents”

arXiv:2510.10931v3 Announce Type: replace Abstract: While reinforcement learning (RL) enhances their ability to plan and reason across retrieval steps, we identify a critical failure mode in this sett

arXiv:2610.11942v1 Announce Type: new Abstract: A login screen that appears after a tap on Sign in is expected; the same screen after a tap on View order is an attack. For GUI agents, safety is theref

arXiv:2610.11781v1 Announce Type: new Abstract: Interactive agents can turn experience into reusable skills, yet existing self-evolving skill libraries primarily improve by accumulating new knowledge.

arXiv:2610.11188v1 Announce Type: new Abstract: Persistent memory can improve personalization in LLM agents but can also induce sycophancy and cross-domain leakage. We distinguish two governance decis

arXiv:2610.09590v1 Announce Type: cross Abstract: Long-running autonomous agents must reuse accumulated reasoning experience without allowing explicit historical memory and LLM context to grow indefin

arXiv:2610.04432v1 Announce Type: new Abstract: Building interactive simulators from real-world observations is a promising way to scale embodied data, but current pipelines still rely heavily on manu

arXiv:2610.02361v1 Announce Type: cross Abstract: Large language model (LLM)-driven evolutionary search is a powerful paradigm for automated program and algorithm discovery, yet existing systems are l

arXiv:2610.00759v1 Announce Type: cross Abstract: Cyber attack agents are typically trained and evaluated within a single simulator, making it unclear whether learned policies transfer beyond the envi

arXiv:2610.00093v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit strong general capabilities, yet their parameters typically remain fixed after deployment, limiting learning from

arXiv:2610.01278v1 Announce Type: new Abstract: Alzheimer's disease (AD) diagnosis requires sequential evidence acquisition under heterogeneous test costs and patient burden. Fixed-modality predictors

arXiv:2609.39102v1 Announce Type: cross Abstract: Self-evolving search agents build their own training curricula by jointly optimizing a proposer that generates questions and a solver that answers the

arXiv:2609.40118v1 Announce Type: new Abstract: As LLM capabilities advance, agents are tackling increasingly complex tasks over longer horizons. Their growing interaction histories make memory compac

arXiv:2609.32868v2 Announce Type: replace-cross Abstract: Traditional scientific computing requires researchers to translate computational intent into environment configuration, resource requests, and

arXiv:2609.37453v1 Announce Type: new Abstract: When a user changes their mind partway through a task, an agent that has already split the task into sub-goals and paid for tool calls must decide, per

arXiv:2609.35911v1 Announce Type: cross Abstract: Adapting large language model agents during deployment requires not only retaining past experience, but also turning new observations into timely guid

arXiv:2609.37111v1 Announce Type: new Abstract: Long-horizon LLM agents require reinforcement learning methods that can assign credit to intermediate decisions under sparse and delayed rewards. Existi

arXiv:2609.35381v2 Announce Type: replace-cross Abstract: Many Model Context Protocol (MCP) servers wrap web APIs built for human developers, and their error messages tell the reader to run a command,

arXiv:2609.36130v1 Announce Type: new Abstract: Long-running LLM agents compress past interactions into persistent memories that may be reused as premises for later tasks. This creates a distinct deri

arXiv:2507.02735v4 Announce Type: replace-cross Abstract: Prompt injection attacks, where untrusted data contains an injected prompt to manipulate the system, have been listed as the top security thre

arXiv:2609.36923v1 Announce Type: new Abstract: Existing reactive Graphical User Interface (GUI) agents often fail in long-horizon, dynamic scenarios, where unexpected disturbances trigger attention-d

arXiv:2609.36939v1 Announce Type: new Abstract: GUI agents automate tasks on digital devices by grounding language instructions in visual interfaces. Existing group-relative reinforcement learning imp

arXiv:2609.37196v1 Announce Type: cross Abstract: Tool-using LLM agents remain vulnerable to indirect prompt injection because trusted instructions and untrusted observations share one context, allowi

arXiv:2609.35808v1 Announce Type: cross Abstract: Experience reuse can reduce repeated exploration in embodied agents, but a trajectory that succeeded previously may be unsuitable for the current exec

arXiv:2609.31675v1 Announce Type: new Abstract: We introduce the Active Causal Discovery Benchmark (ACDB), an SCM-grounded environment for evaluating whether LLM agents recover causal graph structure

arXiv:2609.33391v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) often relies on sparse outcome rewards, providing coarse supervision for long-horizon agents. On-p

arXiv:2609.33875v1 Announce Type: cross Abstract: Outcome-only reinforcement learning gives software engineering (SWE) agents a terminal success signal but little direct guidance about intermediate de

arXiv:2609.34838v1 Announce Type: new Abstract: On-policy distillation (OPD) trains student agents through teacher supervision on their own interactions with an environment. However, in asynchronous m

arXiv:2609.34358v1 Announce Type: new Abstract: In agentic AI systems, frozen foundation models are increasingly deployed as closed-weight API endpoints, making downstream adaptation possible only thr

arXiv:2609.34064v1 Announce Type: new Abstract: Reinforcement learning (RL) has become an effective post-training paradigm for long-horizon large language model (LLM) agents. However, we find that the

arXiv:2609.35515v1 Announce Type: cross Abstract: Scientific discovery requires not only recovering mathematical laws that describe observable behavior, but also identifying the mechanisms that genera

arXiv:2609.33848v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly undertake extreme-long (xlong) horizon tasks, where a single execution can span hours, hundreds of model-

arXiv:2609.32416v1 Announce Type: cross Abstract: Code-as-Policy agents accomplish long-horizon embodied tasks by generating and executing code, yet continually improving them with teachers that are s

arXiv:2609.35319v1 Announce Type: new Abstract: On-policy distillation (OPD) trains a student on its own trajectories with dense teacher supervision. Recent work on OPD for multi-turn autonomous agent

arXiv:2609.32677v1 Announce Type: cross Abstract: Self-improving agents increasingly rely on proxy verifiers to choose policy updates, yet deployment can change the world in which those updates are ev

arXiv:2609.30657v1 Announce Type: cross Abstract: Large language model email assistants are particularly vulnerable to indirect prompt injection because untrusted email content can be retrieved into t

arXiv:2609.28973v1 Announce Type: new Abstract: Physical changes or sensing errors can invalidate embodied agents' task-relevant beliefs. AquaMend compares re-probing, rollback, and supported continua

arXiv:2609.28585v1 Announce Type: cross Abstract: Multi-step tool-calling LLM agents rely on host runtimes to preserve state across turns. When a runtime carries an external tool return into later mod

arXiv:2609.29543v1 Announce Type: new Abstract: Agent skills bundle procedural guidance with execution conditions governing authority, user consent, and live environment state. When model capabilities

arXiv:2606.15912v2 Announce Type: replace-cross Abstract: Multi-turn agents that plan, invoke tools, and interact with environments offer a promising paradigm for solving complex tasks, yet their capa

arXiv:2609.25192v1 Announce Type: new Abstract: Financial search is a highly demanding task for LLM agents, requiring not only a correct final answer but also temporally valid information retrieval, a

arXiv:2609.25636v1 Announce Type: new Abstract: Modern embodied agents achieve impressive success rates, yet their actual instruction-following ability is far weaker than these numbers suggest. We tra

arXiv:2609.22818v1 Announce Type: cross Abstract: Memory-poisoning defenses for LLM agents are typically evaluated by their ability to prevent attacks. However, the traffic they process is rarely adve

arXiv:2609.22222v1 Announce Type: cross Abstract: Large language models can generate executable data-analysis code, but successful execution is not equivalent to a valid official-statistics result. Th

arXiv:2609.22510v1 Announce Type: cross Abstract: As LLM applications integrate with external tools, they are increasingly exposed to indirect prompt injection (IPI), where adversarial instructions ar

arXiv:2609.24259v1 Announce Type: new Abstract: The effectiveness of agent memory ultimately depends on whether the underlying LLM gives each memory in context an appropriate degree of influence over

arXiv:2609.24983v1 Announce Type: new Abstract: We present onPanda, an interactive tool for efficiently annotating LLM alignment data and agent trajectories. onPanda adopts token-level correction as i