Tools as Continuous Flow for Evolving Agentic Reasoning

arXiv:2605.07339 · cs.AI · Submitted 2026-05-08 · Read on arXiv

cs.AI

Submitted: 2026-05-08

Updated: 2026-10-10

Code: https://github.com/ssy166/FlowPlan

Terminology

• Agentic Reasoning — 3× in this paper · explained in 1 episode
• Generalization — 2× in this paper · explained in 13 episodes
• Adaptability — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Flow Matching — 1× in this paper · explained in 10 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Robustness — 1× in this paper · explained in 5 episodes

Sources

• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• A Review of Prominent Paradigms for LLM-Based Agents: Tool Use (Including RAG), Planning, and Feedback Learning
• LLM With Tools: A Survey
• Self-Organizing Agent Network for LLM-based Workflow Automation
• Agent-S: LLM Agentic workflow to automate Standard Operating Procedures
• OdysseyBench: Evaluating LLM Agents on Long-Horizon Complex Office Application Workflows
• ToolRL: Reward is All Tool Learning Needs
• Enhancing LLM-Based Agents via Global Planning and Hierarchical Execution
• Agents of Change: Self-Evolving LLM Agents for Strategic Planning
• Routine: A Structural Planning Framework for LLM Agent System in Enterprise
• Agent-R1: A Unified and Modular Framework for Agentic Reinforcement Learning
• SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?
• Fine-Tuning and Prompt Engineering of LLMs, for the Creation of Multi-Agent AI for Addressing Sustainable Protein Production Challenges
• Chain-of-thought obfuscation learned from output supervision can generalise to unseen tasks
• CATCH: A Novel Data Synthesis Framework for High Therapy Fidelity and Memory-Driven Planning Chain of Thought in AI Counseling
• From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review
• AutoTIR: Autonomous Tools Integrated Reasoning via Reinforcement Learning
• In-the-Flow Agentic System Optimization for Effective Planning and Tool Use
• FlowAgent: Achieving Compliance and Flexibility for Workflow Agents
• Coarse-to-Fine Grounded Memory for LLM Agent Planning

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems