Agents
Reducing Cognitive Overhead in Tool Use via Multi-Small-Agent Reinforcement Learning
arXiv:2508.08882v5 Announce Type: replace Abstract: Recent advances in multi-agent systems highlight the potential of specialized small agents that collaborate via division of labor. Existing tool-int
arXiv:2508.08882v5 Announce Type: replace Abstract: Recent advances in multi-agent systems highlight the potential of specialized small agents that collaborate via division of labor. Existing tool-integrated reasoning systems, however, often follow a single-agent paradigm in which one large model interleaves long-horizon reasoning with precise tool operations, leading to cognitive-load interference and unstable coordination. We present MSARL, a Multi-Small-Agent Reinforcement Learning framework that explicitly decouples reasoning from tool use. In MSARL, a Reasoning Agent decomposes problems and plans tool invocations, while multiple Tool Agents specialize in specific external tools, each trained via a combination of imitation learning and reinforcement learning with role-specific rewards. On mathematical problem solving with code execution, MSARL significantly improves reasoning stability and final-answer accuracy over single-agent baselines. Moreover, the architecture generalizes to diverse tool-use tasks, demonstrating that cognitive-role decoupling with small agents is a scalable blueprint for multi-agent AI design.
Related
- UI-AGILE: Advancing GUI Agents with Effective Reinforcement Learning and Precise Inference-Time Grounding
- Inference-Time Graph Engineering for Multi-Agent LLM Workflows
- Group-Aware Coordination Graph for Multi-Agent Reinforcement Learning
- Agent Primitives: Reusable Latent Building Blocks for Multi-Agent Systems
Source: arXiv cs.AI | 2026-10-02