Imagen de Portada

Tool Attention Is All You Need: Dynamic Tool Gating and Lazy Schema Loading for Eliminating the MCP/Tools Tax in Scalable Agentic Workflows

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Sadani, Anuj, Kumar, Deepak
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Locale-Conditioned Few-Shot Prompting Mitigates Demonstration Regurgitation in On-Device PII Substitution with Small Language Models
por: Sadani, Anuj, et al.
Publicado: (2026)

Are Tools All We Need? Unveiling the Tool-Use Tax in LLM Agents
por: Zhang, Kaituo, et al.
Publicado: (2026)

ToolRL: Reward is All Tool Learning Needs
por: Qian, Cheng, et al.
Publicado: (2025)

Dynamic ReAct: Scalable Tool Selection for Large-Scale MCP Environments
por: Gaurav, Nishant, et al.
Publicado: (2025)

MCP-ITP: An Automated Framework for Implicit Tool Poisoning in MCP
por: Li, Ruiqi, et al.
Publicado: (2026)

GeoMind: An Agentic Workflow for Lithology Classification with Reasoned Tool Invocation
por: Zhou, Yitong, et al.
Publicado: (2026)

TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments
por: Sakizli, Furkan
Publicado: (2026)

ToolBrain: A Flexible Reinforcement Learning Framework for Agentic Tools
por: Le, Quy Minh, et al.
Publicado: (2025)

MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools
por: Wang, Wenhao, et al.
Publicado: (2025)

Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets
por: Sakizli, Furkan
Publicado: (2026)

Semantic Tool Discovery for Large Language Models: A Vector-Based Approach to MCP Tool Selection
por: Mudunuri, Sarat, et al.
Publicado: (2026)

ToolRM: Towards Agentic Tool-Use Reward Modeling
por: Li, Renhao, et al.
Publicado: (2025)

ToolScope: An Agentic Framework for Vision-Guided and Long-Horizon Tool Use
por: Deng, Mengjie, et al.
Publicado: (2025)

Atomix: Timely, Transactional Tool Use for Reliable Agentic Workflows
por: Mohammadi, Bardia, et al.
Publicado: (2026)

Dynamic System Instructions and Tool Exposure for Efficient Agentic LLMs
por: Franko, Uria
Publicado: (2025)

LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?
por: Mo, Guozhao, et al.
Publicado: (2025)

MCP-Zero: Active Tool Discovery for Autonomous LLM Agents
por: Fei, Xiang, et al.
Publicado: (2025)

ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandbox
por: Li, Yuanyang, et al.
Publicado: (2026)

MCP-Atlas: A Large-Scale Benchmark for Tool-Use Competency with Real MCP Servers
por: Bandi, Chaithanya, et al.
Publicado: (2026)

Schema First Tool APIs for LLM Agents: A Controlled Study of Tool Misuse, Recovery, and Budgeted Performance
por: Sigdel, Akshey, et al.
Publicado: (2026)

TOUCAN: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments
por: Xu, Zhangchen, et al.
Publicado: (2025)

Transduction is All You Need for Structured Data Workflows
por: Gliozzo, Alfio, et al.
Publicado: (2025)

Attention is All You Need Until You Need Retention
por: Yaslioglu, M. Murat
Publicado: (2025)

Control Plane as a Tool: A Scalable Design Pattern for Agentic AI Systems
por: Kandasamy, Sivasathivel
Publicado: (2025)

Tools as Continuous Flow for Evolving Agentic Reasoning
por: Huang, Tairan, et al.
Publicado: (2026)

ToolGate: Contract-Grounded and Verified Tool Execution for LLMs
por: Liu, Yanming, et al.
Publicado: (2026)

PyVision: Agentic Vision with Dynamic Tooling
por: Zhao, Shitian, et al.
Publicado: (2025)

Talk Freely, Execute Strictly: Schema-Gated Agentic AI for Flexible and Reproducible Scientific Workflows
por: Strickland, Joel, et al.
Publicado: (2026)

Quantitative Certification of Agentic Tool Selection
por: Yeon, Jehyeok, et al.
Publicado: (2025)

MCP-AgentBench: Evaluating Real-World Language Agent Performance with MCP-Mediated Tools
por: Guo, Zikang, et al.
Publicado: (2025)

ToolMisuseBench: An Offline Deterministic Benchmark for Tool Misuse and Recovery in Agentic Systems
por: Sigdel, Akshey, et al.
Publicado: (2026)

AdapTools: Adaptive Tool-based Indirect Prompt Injection Attacks on Agentic LLMs
por: Wang, Che, et al.
Publicado: (2026)

ACE-Router: Generalizing History-Aware Routing from MCP Tools to the Agent Web
por: Yao, Zhiyuan, et al.
Publicado: (2026)

ParaView-MCP: An Autonomous Visualization Agent with Direct Tool Use
por: Liu, Shusen, et al.
Publicado: (2025)

ToolLibGen: Scalable Automatic Tool Creation and Aggregation for LLM Reasoning
por: Yue, Murong, et al.
Publicado: (2025)

GTA-2: Benchmarking General Tool Agents from Atomic Tool-Use to Open-Ended Workflows
por: Wang, Jize, et al.
Publicado: (2026)

Secure Tool Manifest and Digital Signing Solution for Verifiable MCP and LLM Pipelines
por: Jamshidi, Saeid, et al.
Publicado: (2026)

From Tool Orchestration to Code Execution: A Study of MCP Design Choices
por: Felendler, Yuval, et al.
Publicado: (2026)

VerlTool: Towards Holistic Agentic Reinforcement Learning with Tool Use
por: Jiang, Dongfu, et al.
Publicado: (2025)

Element-wise Attention Is All You Need
por: Feng, Guoxin
Publicado: (2025)