Tool-Using Agents

Follow this topic Ask about this topic

Tool-using agents are language-model-based systems that choose and invoke external tools—such as search, code, databases, or graphical interfaces—and use the results to pursue goals over multiple steps. Research develops tool selection and retrieval, planning and decomposition, memory and reusable skills, and training with supervised trajectories or reinforcement learning; runtime harnesses add state tracking, verification, permissions, and recovery. Evaluation is shifting from isolated function calls and final answers toward stateful workflows, measuring task completion alongside cost, policy adherence, robustness to errors, and intermediate actions across web, terminal, office, and domain-specific tasks. A central tension is that broader autonomy and longer tool chains expand capability but also amplify grounding errors, stale state, unsafe side effects, and hidden tool failures, so reliable execution and calibrated stopping remain open problems.

Latest papers

289 papers54 papers in the last four weeks, up 74% on the four weeks before. 0.3% of all new papers.

Agent Meltdowns: The Road to Hell Is Paved with Helpful Agents
May 18, 2026Rishi Jha, Harold Triedman, Arkaprabha Bhattacharya +1Agent Failure Analysis AI Agent Reliability
Department of Computer Science Cornell University
TOBench: A Task-Oriented Omni-Modal Benchmark for Real-World Tool-Using Agents
May 16, 2026Zhiqiang Liu, Wenhui Dong, Yilang Tan +3Tool-Use Evaluation AI Agent Benchmarks
Nanjing University · Huazhong University of Science and Technology · Southwest Jiaotong University +1
CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?
May 15, 2026Haolin Chen, Deon Metelski, Leon Qi +30Healthcare Long-Horizon Agent Evaluation
Johns Hopkins Medicine · Wellstar Health System · Stanford University +16
R2V Agent: Teaching SLMs When to Ask for Help
May 15, 2026Raghu Vamshi Hemadri, Humaira Firdowse Mohammed, Rishabh Maheshwary +5AI Agent Reliability Tool-Augmented Language Model Agents
New York University Tandon School of Engineering · University of California, San Diego · ServiceNow Research +3
ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing
May 15, 2026Lijie Ding, Changwoo DoAgentic Workflows AI Agent Benchmarks
Neutron Scattering Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA
Multi-Agentic Approach for History Matching of Oil Reservoirs
May 14, 2026Linar Samigullin, Sergei Shumilin, Evgeny BurnaevMulti-Agent Orchestration LLM Agent Orchestration
Skoltech, AI Center, Moscow, Russia · AIRI, Moscow, Russia
ChromaFlow: A Negative Ablation Study of Orchestration Overhead in Tool-Augmented Agent Evaluation
May 13, 2026Tarun MittalAI Agent Reliability LLM Agent Orchestration
Octave-X
CRANE: Constrained Reasoning Injection for Code Agents via Nullspace Editing
May 13, 2026Mingzhi Zhu, Michele Merler, Raju Pavuluri +1Software Engineering Agents Agentic Code Generation
Rensselaer Polytechnic Institute Troy, NY 12180 · IBM Research Yorktown Heights, NY 10598
RS-Claw: Progressive Active Tool Exploration via Hierarchical Skill Trees for Remote Sensing Agents
May 13, 2026Liangtian Liu, Zeyuan Wang, Ziyu Li +8Remote Sensing Image Understanding Tool-Using Agents
School of Geosciences and Info-Physics, Central South University, Changsha 410083, China · School of Resources and Environment, University of Electronic Science and Technology of China, Xian 710071, China · School of Earth Sciences and Spatial Information Engineering, Hunan University of Science and Technology, Xiangtan 411201, China +1
Speculative Interaction Agents: Building Real-Time Agents with Asynchronous I/O and Speculative Tool Calling
May 13, 2026Coleman Hooper, Minwoo Kang, Suhong Moon +7Tool-Using Agents LLM Agents
University of California, Berkeley · ICSI · LBNL
No Action Without a NOD: A Heterogeneous Multi-Agent Architecture for Reliable Service Agents
May 12, 2026Zixu Yang, Hang Zheng, Nan Jiang +5AI Agent Reliability Multi-Agent LLM Systems
1X-LANCE Lab, School of Computer Science, Shanghai Jiao Tong University, Shanghai, China · 2Shanghai Innovation Institution, Shanghai, China · 3Jiangsu Key Lab of Language Computing, Suzhou, China
When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents
May 12, 2026Xiaolin Zhou, Aojie Yuan, Zheng Luo +12Domain Randomization AI Agent Benchmarks
Arizona State University · University of Southern California · University of Pennsylvania +2
Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations
May 12, 2026Junjue Wang, Weihao Xuan, Heli Qi +7LLM Agent Evaluation Geospatial Reasoning
The University of Tokyo · RIKEN AIP · Waseda University +1
The Semantic Training Gap: Ontology-Grounded Tool Architectures for Industrial AI Agent Systems
May 11, 2026Grama ChethanLLM Grounding LLM Hallucination Mitigation
Siemens Digital Industries Software, Plano, TX, United States
ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandbox
May 11, 2026Yuanyang Li, Xue Yang, Longyue Wang +2LLM Agent Evaluation AI Agent Benchmarks
1Zhejiang University · 2Zhejiang Lab · 3Alibaba Group
Agent-First Tool API: A Semantic Interface Paradigm for Enterprise AI Agent Systems
May 11, 2026Kai PanTool-Using Agents AI Agent Governance
A2A Lab
Security Risks in Tool-Enabled AI Agents: A Systematic Analysis of Privileged Execution Environments
May 10, 2026Hardik GoelAI Agent Security Tool-Using Agents
Microsoft Corporation, USA
CIVeX: Causal Intervention Verification for Language Agents
May 9, 2026Fabio RovaiStructural Causal Models Tool-Using Agents
The Tesseract Academy
Octopus Protocol: One-Shot Hardware Discovery and Control for AI Agents via Infrastructure-as-Prompts
May 9, 2026Quilee Simeon, Justin M. Wei, Yile FanRobotics AI Control
MIT · Harvard
AI-Care: A Conversational Agentic System for Task Coordination in Alzheimer's Disease Care
May 8, 2026Preyash Yadav, Michelle Cohn, Priyanka Koppolu +5Alzheimer's Disease Tool-Using Agents
Department of Computer Science University of California, Davis Davis, USA · Department of Linguistics University of California, Davis Davis, USA · Department of Neurology University of California, Davis Sacramento, USA +1
Do Agents Need to Plan Step-by-Step? Rethinking Planning Horizon in Data-Centric Tool Calling
May 8, 2026Naoki Otani, Nikita Bhutani, Hannah Kim +2Long-Horizon Planning Tool-Using Agents
Megagon Labs Mountain View, California, USA
AgentEscapeBench: Evaluating Out-of-Domain Tool-Grounded Reasoning in LLM Agents
May 8, 2026Zhengkang Guo, Yiyang Li, Lin Qiu +7LLM Agent Evaluation Long-Horizon Agent Evaluation
1Fudan University · 2Meituan Longcat Team
Beyond the Black Box: Interpretability of Agentic AI Tool Use
May 7, 2026Hariom Tatsat, Ariye ShaterAI Agent Reliability Mechanistic Interpretability
Quantitative Analytics, Barclays
MANTRA: Synthesizing SMT-Validated Compliance Benchmarks for Tool-Using LLM Agents
May 7, 2026Ashwani Anand, Ivi Chatzi, Ritam Raha +1Computer-Use Agent Benchmarks AI Agent Reliability
Max Planck Institute for Software Systems, Kaiserslautern, Germany
LongSeeker: Elastic Context Orchestration for Long-Horizon Search Agents
May 6, 2026Yijun Lu, Rui Ye, Yuwen Du +3Web Search Agents Tool-Using Agents
Shanghai Jiao Tong University
Terminus-4B: Can a Smaller Model Replace Frontier LLMs at Agentic Execution Tasks?
May 4, 2026Spandan Garg, Vikram Nitin, Yufan HuangAI Coding Agents Small Language Models
Microsoft USA
Beyond State Machines: Executing Network Procedures with Agentic Tool-Calling Sequences
May 4, 2026Purna Sai Garigipati, Onur Ayan, Kishor Chandra Joshi +1LLM Agent Orchestration Tool-Using Agents
Heisenberg Research Center, Huawei Technologies Duesseldorf GmbH, 80992 Munich, Germany · Eindhoven University of Technology, Eindhoven, The Netherlands
TADI: Tool-Augmented Drilling Intelligence via Agentic LLM Orchestration over Heterogeneous Wellsite Data
Apr 30, 2026Rong LuLLM Agent Orchestration Tool-Using Agents
Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents
Apr 29, 2026Anh Ta, Junjie Zhu, Shahin ShayandehMulti-Agent LLM Systems Tool-Use Evaluation
Apple
Bidirectional Semantic Complementary Tool Retrieval for Remote Sensing Agents
Apr 29, 2026Zeyuan Wang, Dongyang Hou, Cheng Yang +10Remote Sensing Tool-Using Agents
School of Geosciences and Info-Physics, Central South University, Changsha 410083, China. · School of Mechanical and Electrical Engineering, Central South University, Changsha 410083, China. · Hunan Key Laboratory of Land Resources Evaluation and Utilization, Hunan Provincial Institute of Land and Resources Planning, Changsha 410013, China.

May 18, 2026J/K move · Enter open · S save

Agent Meltdowns: The Road to Hell Is Paved with Helpful Agents

Rishi Jha, Harold Triedman, Arkaprabha Bhattacharya, Vitaly Shmatikov

Department of Computer Science Cornell University

Agent Failure Analysis AI Agent Reliability LLM Agent Evaluation AI Agent Safety Tool-Using Agents

Open paper AskPDF arXiv

Agents operating with computer and Web use inevitably encounter errors: inaccessible webpages, missing files, local and remote misconfigurations, etc. These errors do not thwart agents based on state-of-the-art models. They helpfully continue to look for ways to complete their tasks. We introduce, characterize, and measure a new type of agent failure we call accidental meltdown: unsafe or harmful behavior in response to a benign environmental error, in the absence of any adversarial inputs. Because meltdowns are not captured by the existing reliability or safety benchmarks, we develop a taxonomy of meltdown behaviors. We then implement an agent-agnostic infrastructure for injecting simulated local and remote errors into the rollout environment and use it to systematically evaluate agent systems powered by GPT, Grok, and Gemini. Our evaluation demonstrates that meltdowns (e.g., conducting unauthorized reconnaissance or subverting access control) of varying severity and success occur in 64.7% of agent rollouts that encounter simulated errors, spanning all combinations of agent system, backing model, and error type. In over half of these meltdowns, unsafe behaviors are not reported to the user. Comparing behaviors of the same agents with and without errors, we find that exploration in response to errors is correlated with unsafe and harmful behavior.