Image de couverture de livre

Are Tools All We Need? Unveiling the Tool-Use Tax in LLM Agents

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhang, Kaituo, Xiong, Zhen, Zhong, Mingyu, Jiang, Zhimeng, Yuan, Zhouyuan, Li, Zhecheng, Lin, Ying
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Cleansing the Artificial Mind: A Self-Reflective Detoxification Framework for Large Language Models
par: Zhang, Kaituo, et autres
Publié: (2026)

Tool Attention Is All You Need: Dynamic Tool Gating and Lazy Schema Loading for Eliminating the MCP/Tools Tax in Scalable Agentic Workflows
par: Sadani, Anuj, et autres
Publié: (2026)

Learning to Rewrite Tool Descriptions for Reliable LLM-Agent Tool Use
par: Guo, Ruocheng, et autres
Publié: (2026)

UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents
par: Liang, Yijuan, et autres
Publié: (2026)

FinToolBench: Evaluating LLM Agents for Real-World Financial Tool Use
par: Lu, Jiaxuan, et autres
Publié: (2026)

ToolRL: Reward is All Tool Learning Needs
par: Qian, Cheng, et autres
Publié: (2025)

ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
par: Feng, Jiazhan, et autres
Publié: (2025)

Utility-Guided Agent Orchestration for Efficient LLM Tool Use
par: Liu, Boyan, et autres
Publié: (2026)

When Does Memory Help Multi-Trajectory Inference for Tool-Use LLM Agents?
par: Li, Xinzhe, et autres
Publié: (2026)

The LLM Data Auditor: A Metric-oriented Survey on Quality and Trustworthiness in Evaluating Synthetic Data
par: Zhang, Kaituo, et autres
Publié: (2026)

ToolCUA: Towards Optimal GUI-Tool Path Orchestration for Computer Use Agents
par: Hu, Xuhao, et autres
Publié: (2026)

Benchmarking LLM Tool-Use in the Wild
par: Yu, Peijie, et autres
Publié: (2026)

EvoTool: Self-Evolving Tool-Use Policy Optimization in LLM Agents via Blame-Aware Mutation and Diversity-Aware Selection
par: Yang, Shuo, et autres
Publié: (2026)

Reward Hacking Benchmark: Measuring Exploits in LLM Agents with Tool Use
par: Thaman, Kunvar
Publié: (2026)

SoK: Agentic Skills -- Beyond Tool Use in LLM Agents
par: Jiang, Yanna, et autres
Publié: (2026)

Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Use
par: Cheng, Yize, et autres
Publié: (2026)

ToolTweak: An Attack on Tool Selection in LLM-based Agents
par: Sneh, Jonathan, et autres
Publié: (2025)

MMedAgent: Learning to Use Medical Tools with Multi-modal Agent
par: Li, Binxu, et autres
Publié: (2024)

Butterfly Effects in Toolchains: A Comprehensive Analysis of Failed Parameter Filling in LLM Tool-Agent Systems
par: Xiong, Qian, et autres
Publié: (2025)

ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents
par: Li, Dawei, et autres
Publié: (2026)

Budget-Aware Tool-Use Enables Effective Agent Scaling
par: Liu, Tengxiao, et autres
Publié: (2025)

AgentEscapeBench: Evaluating Out-of-Domain Tool-Grounded Reasoning in LLM Agents
par: Guo, Zhengkang, et autres
Publié: (2026)

ReTool-Video: Recursive Tool-Using Video Agents with Meta-Augmented Tool Grounding
par: Liu, Xiao, et autres
Publié: (2026)

Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover
par: Chung, Jui-Hui, et autres
Publié: (2026)

From Imperative to Declarative: Towards LLM-friendly OS Interfaces for Boosted Computer-Use Agents
par: Wang, Yuan, et autres
Publié: (2025)

GTA-2: Benchmarking General Tool Agents from Atomic Tool-Use to Open-Ended Workflows
par: Wang, Jize, et autres
Publié: (2026)

ToolRM: Towards Agentic Tool-Use Reward Modeling
par: Li, Renhao, et autres
Publié: (2025)

Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning
par: Xu, Ningning, et autres
Publié: (2025)

Experience-Evolving Multi-Turn Tool-Use Agent with Hybrid Episodic-Procedural Memory
par: Li, Sijia, et autres
Publié: (2025)

Your LLM Agent Can Leak Your Data: Data Exfiltration via Backdoored Tool Use
par: Zhang, Wuyang, et autres
Publié: (2026)

Attractive Metadata Attack: Inducing LLM Agents to Invoke Malicious Tools
par: Mo, Kanghua, et autres
Publié: (2025)

Enhancing LLM Character-Level Manipulation via Divide and Conquer
par: Xiong, Zhen, et autres
Publié: (2025)

Agents Are All You Need for LLM Unlearning
par: Sanyal, Debdeep, et autres
Publié: (2025)

Proof-of-Use: Mitigating Tool-Call Hacking in Deep Research Agents
par: Ma, SHengjie, et autres
Publié: (2025)

FamilyTool: A Multi-hop Personalized Tool Use Benchmark
par: Wang, Yuxin, et autres
Publié: (2025)

CoVe: Training Interactive Tool-Use Agents via Constraint-Guided Verification
par: Chen, Jinpeng, et autres
Publié: (2026)

MCPAgentBench: A Real-world Task Benchmark for Evaluating LLM Agent MCP Tool Use
par: Liu, Wenrui, et autres
Publié: (2025)

Were RNNs All We Needed?
par: Feng, Leo, et autres
Publié: (2024)

SearchSkill: Teaching LLMs to Use Search Tools with Evolving Skill Banks
par: Hu, Jinchao, et autres
Publié: (2026)

AutoTool: Efficient Tool Selection for Large Language Model Agents
par: Jia, Jingyi, et autres
Publié: (2025)