认知内核Pro：深度研究智能体与智能体基础模型训练框架

摘要

通用人工智能（General AI）代理正日益被视为下一代人工智能的基础框架，它们能够实现复杂的推理、网络交互、编程以及自主研究能力。然而，当前的代理系统要么是闭源的，要么严重依赖多种付费API和专有工具，这限制了研究社区的可访问性和可复现性。在本研究中，我们推出了Cognitive Kernel-Pro，这是一个完全开源且（在最大程度上）免费的多模块代理框架，旨在普及高级AI代理的开发和评估。在Cognitive Kernel-Pro中，我们系统地研究了高质量训练数据的整理，重点关注在四个关键领域——网络、文件、代码和通用推理——中构建查询、轨迹和可验证答案的方法。此外，我们探索了代理测试时反思和投票的新策略，以增强代理的鲁棒性和性能。我们在GAIA基准上对Cognitive Kernel-Pro进行了评估，取得了开源和免费代理中的最先进成果。值得注意的是，我们的8B参数开源模型超越了之前领先的系统，如WebDancer和WebSailor，为可访问的高能力AI代理树立了新的性能标准。代码已发布于https://github.com/Tencent/CognitiveKernel-Pro。

English

General AI Agents are increasingly recognized as foundational frameworks for the next generation of artificial intelligence, enabling complex reasoning, web interaction, coding, and autonomous research capabilities. However, current agent systems are either closed-source or heavily reliant on a variety of paid APIs and proprietary tools, limiting accessibility and reproducibility for the research community. In this work, we present Cognitive Kernel-Pro, a fully open-source and (to the maximum extent) free multi-module agent framework designed to democratize the development and evaluation of advanced AI agents. Within Cognitive Kernel-Pro, we systematically investigate the curation of high-quality training data for Agent Foundation Models, focusing on the construction of queries, trajectories, and verifiable answers across four key domains: web, file, code, and general reasoning. Furthermore, we explore novel strategies for agent test-time reflection and voting to enhance agent robustness and performance. We evaluate Cognitive Kernel-Pro on GAIA, achieving state-of-the-art results among open-source and free agents. Notably, our 8B-parameter open-source model surpasses previous leading systems such as WebDancer and WebSailor, establishing a new performance standard for accessible, high-capability AI agents. Code is available at https://github.com/Tencent/CognitiveKernel-Pro

认知内核Pro：深度研究智能体与智能体基础模型训练框架

Cognitive Kernel-Pro: A Framework for Deep Research Agents and Agent Foundation Models Training

摘要

Support