Search: “tool-use”

arXiv:2606.22936v2 Announce Type: replace Abstract: Tool-using LLM agents make early choices about what to search for, which evidence to trust, and what to do next. A wrong early choice can steer ever

arXiv:2608.29128v1 Announce Type: new Abstract: Tool-using agents are commonly evaluated by a single bit: whether an end-to-end workflow completed. This metric fails to distinguish failures that matte

arXiv:2608.18351v1 Announce Type: cross Abstract: Tool-using large language-model agents can complete a task while exercising authority that the user did not grant or the task does not need, causing e

arXiv:2512.21414v2 Announce Type: replace Abstract: Recent tool-use frameworks powered by vision-language models (VLMs) improve image understanding by grounding model predictions with specialized tool

arXiv:2510.26020v2 Announce Type: replace Abstract: Multi-tool-integrated reasoning enables LLM-empowered tool-use agents to solve complex tasks by interleaving natural-language reasoning with calls t

arXiv:2610.06401v1 Announce Type: cross Abstract: Tool-using language-model agents are vulnerable to indirect prompt injection because they must act on untrusted external content. Existing training-ti

arXiv:2609.13060v1 Announce Type: new Abstract: Diffusion Language Models (DLMs) offer promising parallel generation capabilities but lag behind autoregressive models in complex reasoning and tool-use

arXiv:2609.06124v1 Announce Type: new Abstract: High-quality multi-turn tool-use data is essential for training agentic models, yet existing data synthesis methods often underrepresent the argument-le

arXiv:2608.29397v1 Announce Type: new Abstract: Tool-use benchmarks generally evaluate whether an agent completes a workflow using appropriate tools and valid arguments. However, feasibility alone is

arXiv:2607.11183v2 Announce Type: replace Abstract: Large language models increasingly operate as tool-using agents, where small format, argument, or function-call errors can invalidate otherwise plau

arXiv:2605.17774v1 Announce Type: new Abstract: Large language models are increasingly used as planning components in agentic systems, but current tool-use pipelines often require full tool schemas to

arXiv:2605.09822v1 Announce Type: cross Abstract: We define Oracle Poisoning, an attack class in which an adversary corrupts a structured knowledge graph that AI agents query at runtime via tool-use p

arXiv:2604.19945v1 Announce Type: new Abstract: In this paper, we investigate the problem of how to effectively master tool-use to solve complex visual reasoning tasks for Multimodal Large Language Mo

arXiv:2604.10261v1 Announce Type: new Abstract: Existing tool-use benchmarks for LLM agents are overwhelmingly linear: our analysis of six benchmarks shows 55 to 100% of instances are simple chains of

arXiv:2610.21045v1 Announce Type: new Abstract: Tool-using language agents must revise prepared actions when evidence changes or authority is revoked. Selective withdrawal means suspending affected ac

arXiv:2610.14346v2 Announce Type: replace Abstract: Enterprise customer-support agents increasingly rely on Large Language Models (LLMs) to execute bounded, tool-using workflows. We study this problem

arXiv:2610.15246v1 Announce Type: new Abstract: Tool-using agents alternate between model generation and tool execution, waiting for each tool call to complete before starting the next generation. Thi

arXiv:2610.13330v1 Announce Type: new Abstract: Designing tool-integrated reasoning agent systems for complex chemical tasks remains a fundamental challenge. Although recent chemical agents have demon

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 m

arXiv:2610.00949v1 Announce Type: new Abstract: Supervised fine-tuning (SFT) on offline agent trajectories is the standard approach for training specialized tool-using agents, but forcing models to im

arXiv:2609.39378v1 Announce Type: new Abstract: Real-world embodied tasks, from everyday activities to professional procedures, require agents to act under physical constraints while tracking evolving

arXiv:2609.38805v1 Announce Type: cross Abstract: LLM agents often admit multiple high-quality solutions to the same task, differing in reasoning structure, tool-use pattern, or interaction trajectory

arXiv:2609.39566v1 Announce Type: new Abstract: Foundation models can serve as clinical agents through tool-use harnesses. However, conventional medical benchmarks assess reasoning over preselected ev

arXiv:2603.19896v2 Announce Type: replace Abstract: Tool-using large language model (LLM) agents often face a fundamental tension between answer quality and execution cost. Fixed workflows are stable

arXiv:2608.24870v1 Announce Type: new Abstract: Group-relative reinforcement learning waits for sibling rollouts of the same prompt, which is costly for long and variable tool-use trajectories. Single

arXiv:2608.23635v1 Announce Type: cross Abstract: Large language models (LLMs) rely on tool calling as a fundamental agent capability, enabling them to invoke external systems and complete tasks beyon

arXiv:2608.14035v1 Announce Type: new Abstract: Recent developments in large language models (LLMs) and tool-using agents encourage people to explore the potential of using agents in chip design. The

arXiv:2608.02876v1 Announce Type: new Abstract: Tool-using agents do not merely consume observations: their actions determine what arrives next. In agentic text-to-SQL, a broad query can spend context

arXiv:2608.02391v1 Announce Type: cross Abstract: Tool-using large language model (LLM) agents produce long, multi-turn trajectories, making gradient-based post-training memory-intensive. Evolution st

arXiv:2512.06716v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly used as the core of agentic systems due to their strong reasoning, planning, and tool-use capabi

arXiv:2607.26977v1 Announce Type: new Abstract: Travel planning is a demanding stress test for tool-using LLM agents: a usable itinerary is a single artifact that must be right along many axes at once

arXiv:2607.20491v1 Announce Type: new Abstract: Standard evaluation benchmarks measure what a tool-using agent decides, not whether it arrives at that decision through the same process each time. We i

arXiv:2606.06976v1 Announce Type: new Abstract: Large language model (LLM)-based agents often make suboptimal tool-use decisions, including unsupported tool invocation and hallucinated direct response

arXiv:2606.00096v1 Announce Type: cross Abstract: Visual agents employ external visual tools within visual chains of thought to incorporate fine-grained evidence. While prior work has mainly studied t

arXiv:2606.02518v1 Announce Type: new Abstract: Fine-grained image classification (FGIC) has broad applications and has attracted significant research attention. In this paper, we explore a novel para

arXiv:2605.26952v1 Announce Type: new Abstract: Agentic reinforcement learning (RL) has proven effective for training LLM-based agents with external tool-use capabilities. However, we identify that ag

arXiv:2605.14084v1 Announce Type: cross Abstract: Code agents must both reason over long-horizon repository state and obey strict tool-use protocols. In paired Instruct/Thinking checkpoints, these cap

arXiv:2511.17085v3 Announce Type: replace-cross Abstract: The deployment of Large Language Models (LLMs) as tool-using agents causes their alignment training to manifest in new ways. Recent work finds

arXiv:2610.19560v1 Announce Type: cross 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. Y

arXiv:2610.16251v1 Announce Type: cross Abstract: Deep-search agents are commonly trained by distilling tool-use trajectories from stronger teacher models. However, structurally complex questions may

arXiv:2610.15153v1 Announce Type: new Abstract: We ask whether a tool-using agent whose sole duty is to keep humans alive can be induced, by delayed prompt injection, to kill. The keep-alive duty is i

arXiv:2610.19638v1 Announce Type: new Abstract: We present J-tau, the first publicly available Japanese benchmark for evaluating tool-using LLM agents in interactive customer-support scenarios. While

arXiv:2610.14774v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly expected to orchestrate diverse tools over long-horizon tasks. Existing agents typically use Direct Tool C

arXiv:2610.11030v1 Announce Type: cross Abstract: Tool-using LLM agents violate the policies they are deployed to enforce, often silently. Prior defenses hand-write rules, query an LLM verifier per ac

Over the past several months, we conducted an empirical post-mortem investigation into 109 autonomous AI agent security incidents (cataloged with 193 falsification criteria across tool-use and multi-a

arXiv:2601.22352v2 Announce Type: replace-cross Abstract: Language model agents often appear capable of self-recovery after failing tool call executions, yet this behavior lacks a formal explanation.