Core Viewpoint - The article highlights the journey of Wu Minghui, the founder of Minglue Technology, emphasizing his technical background, entrepreneurial challenges, and the evolution of his company towards AI-driven solutions, particularly focusing on trust and data credibility in business decision-making. Group 1: Entrepreneurial Journey - Wu Minghui is portrayed as a typical "scholar-type" entrepreneur with a strong technical background, having excelled in mathematics and computer science [1][7] - The company experienced significant ups and downs, including a dramatic downturn where it struggled to pay severance to employees, leading to negative public perception [1][39][46] - After nearly two decades of exploration in the business world, Wu has focused on the core question of what constitutes trustworthy data [3][24] Group 2: Product Development and Innovation - Minglue Technology recently launched the multi-modal foundational model web GUI intelligent agent, Mano, which achieved state-of-the-art performance in international benchmarks [1][2] - The proprietary large model product line, DeepMiner, aims to address the challenge of making AI agents trustworthy, explainable, and traceable in enterprise decision-making [2][68] - DeepMiner is designed to connect credible data sources, enabling businesses to make informed decisions based on reliable data analysis [68][69] Group 3: Strategic Insights and Reflections - Wu reflects on the importance of trust in data and the need for AI to act as a gatekeeper in business decisions [4][66] - The article discusses the strategic errors made during the company's rapid expansion, emphasizing the need for a controlled strategic pace [50][51] - Wu acknowledges the lessons learned from past failures, particularly the necessity of aligning team goals and maintaining trust within the organization [54][57]

Core Viewpoint - Minglue Technology's proprietary GUI model, Mano, has achieved record-breaking SOTA results in the recognized benchmarks Mind2Web and OSWorld, establishing a new paradigm for GUI intelligent agents through innovations in online reinforcement learning and automatic data collection [1][14][23]. Group 1: Performance Achievements - Mano achieved a success rate of 40.1% in the OSWorld-Verified benchmark, surpassing other models such as qwen and GUI-Owl [10][19]. - In the Mind2Web benchmark, Mano demonstrated superior performance across various metrics, including element accuracy and step success rate, significantly outperforming all other SOTA methods [18][15]. - The model's success rate in OSWorld-Verified reached 41.6±0.7%, marking an approximate 7 percentage point improvement over competitors [21][19]. Group 2: Innovations and Methodology - Mano introduces online reinforcement learning as a novel training paradigm in the GUI interaction field, enhancing its performance in dynamic environments [22][23]. - The model's architecture consists of three main components: exploration module, processing flow, and optimization process, which collectively improve its reasoning and adaptability [25][26]. - The automatic data collection method developed by the technical team significantly enhances the efficiency and accuracy of data acquisition, allowing for the generation of high-quality interaction trajectory data [48][49]. Group 3: Market Context and Future Directions - The demand for AI agents is expected to surge by 2025, positioning Mano as a key player in differentiated competition by accessing data sources that other agents cannot reach [59][63]. - Minglue Technology plans to continue exploring areas such as data collection, training integration, and CAPTCHA handling to further optimize Mano for real-world applications [66].