阿特拉斯：協調異構模型與工具實現多領域複雜推理

摘要

大型語言模型（LLMs）與外部工具的整合已顯著擴展了AI代理的能力。然而，隨著LLMs與工具多樣性的同步增長，選擇最優的模型-工具組合已成為高維度優化難題。現有方法通常依賴單一模型或固定工具調用邏輯，未能充分發掘異構模型-工具配對間的效能差異。本文提出ATLAS（自適應工具-LLM對齊與協同調用框架），這是一種用於跨領域複雜推理的雙路徑動態工具調用框架。ATLAS通過雙路徑機制運作：（1）基於無訓練聚類的路由策略，利用經驗先驗實現領域自適應對齊；（2）基於強化學習的多步路由策略，探索自主軌跡以實現分佈外泛化。在15個基準測試上的大量實驗表明，本方法優於GPT-4o等閉源模型，在分佈內任務（+10.1%）和分佈外任務（+13.1%）上均超越現有路由方法。此外，本框架通過協調專業化多模態工具，在視覺推理任務中展現出顯著效能提升。

English

The integration of large language models (LLMs) with external tools has significantly expanded the capabilities of AI agents. However, as the diversity of both LLMs and tools increases, selecting the optimal model-tool combination becomes a high-dimensional optimization challenge. Existing approaches often rely on a single model or fixed tool-calling logic, failing to exploit the performance variations across heterogeneous model-tool pairs. In this paper, we present ATLAS (Adaptive Tool-LLM Alignment and Synergistic Invocation), a dual-path framework for dynamic tool usage in cross-domain complex reasoning. ATLAS operates via a dual-path approach: (1) training-free cluster-based routing that exploits empirical priors for domain-specific alignment, and (2) RL-based multi-step routing that explores autonomous trajectories for out-of-distribution generalization. Extensive experiments across 15 benchmarks demonstrate that our method outperforms closed-source models like GPT-4o, surpassing existing routing methods on both in-distribution (+10.1%) and out-of-distribution (+13.1%) tasks. Furthermore, our framework shows significant gains in visual reasoning by orchestrating specialized multi-modal tools.

阿特拉斯：協調異構模型與工具實現多領域複雜推理

Atlas: Orchestrating Heterogeneous Models and Tools for Multi-Domain Complex Reasoning

摘要

Support