轻量级框架统一动作数据，加速大模型训练。
ActionStudio: A Lightweight Framework for Data and Training of Large Action Models
- 提出统一格式2.0，整合多源智能体轨迹。
- 训练吞吐提升9倍，性能达公开基准最优。
- 适合需要高效训练智能体的大模型研究者。
大型动作模型对实现自主智能体完成复杂任务至关重要，但其训练面临环境多样性和噪声智能体数据复杂的挑战。现有基础设施在可扩展的、针对特定智能体的微调及标准化数据处理方面支持有限。我们提出ActionStudio，一个专为大型动作模型设计的轻量且可扩展的数据与训练框架。该框架采用自研的统一格式2.0，统一多种智能体轨迹；支持多种训练流程，具备优化的多节点分布式设置；集成强大的预处理与实时验证工具。实验表明，ActionStudio相比现有智能体训练框架，吞吐最高提升9倍；基于其训练的模型在公共及真实场景智能体基准测试中表现领先。为推动社区发展，我们开源ActionStudio框架，并发布actionstudio-98k数据集，包含98,000条高质量轨迹。代码见：https://github.com/SalesforceAIResearch/xLAM。
原文摘要 · Abstract (English)
Large Action models are essential for enabling autonomous agents to perform complex tasks. However, training such models remains challenging due to the diversity of agent environments and the complexity of noisy agentic data. Existing infrastructure offers limited support for scalable, agent-specific fine-tuning and standardized agent data processing. We introduce ActionStudio, a lightweight and extensible data and training framework designed for large action models. ActionStudio unifies diverse agent trajectories using our proposed Unified Format 2.0, supports a range of training workflows with optimized multi-node distributed setup, and integrates robust preprocessing and real-time verification tools. ActionStudio demonstrates up to 9x higher throughput compared to existing agentic training frameworks, and our trained models yield top performances across public and realistic agent benchmarks. To support the broader research community, we open-source the ActionStudio framework and release actionstudio-98k, a curated dataset of 98k high-quality trajectories. Code: https://github.com/SalesforceAIResearch/xLAM.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。