Xiaomi

Jiaze Li, Jingyang Chen, Yuxun Qu, Shijie Xu, Zhenrui Lin, Jun-Yan Zhu, Boshen Xu, Wenhui Tan, Pei Fu, Jianzhong Ju, Zhenbo Luo, Jian Luan

We open-source MiMo-VL-Miloco-7B and its quantized variant MiMo-VL-Miloco-7B-GGUF, a pair of home-centric vision-language models that achieve strong performance on both home-scenario understanding and general multimodal reasoning. Built on the MiMo-VL-7B backbone, MiMo-VL-Miloco-7B is specialized for smart-home environments, attaining leading F1 scores on gesture recognition and common home-scenario understanding, while also delivering consistent gains across video benchmarks such as Video-MME, Video-MMMU, and Charades-STA, as well as language understanding benchmarks including MMMU-Pro and MMLU-Pro. In our experiments, MiMo-VL-Miloco-7B outperforms strong closed-source and open-source baselines on home-scenario understanding and several multimodal reasoning benchmarks. To balance specialization and generality, we design a two-stage training pipeline that combines supervised fine-tuning with reinforcement learning based on Group Relative Policy Optimization, leveraging efficient multi-domain data. We further incorporate chain-of-thought supervision and token-budget-aware reasoning, enabling the model to learn knowledge in a data-efficient manner while also performing reasoning efficiently. Our analysis shows that targeted home-scenario training not only enhances activity and gesture understanding, but also improves text-only reasoning with only modest trade-offs on document-centric tasks. Model checkpoints, quantized GGUF weights, and our home-scenario evaluation toolkit are publicly available at https://github.com/XiaoMi/xiaomi-mimo-vl-miloco to support research and deployment in real-world smart-home applications.

Erjia Xiao, Lingfeng Zhang, Yingbo Tang, Hao Cheng, Renjing Xu, Wenbo Ding, Lei Zhou, Long Chen, Hangjun Ye, Xiaoshuai Hao

Li Pei, Caiting Dong

PurposeThe synergy between platform owners and complementors plays a crucial role in the evolution of platform ecosystems. However, there is limited understanding of the correlation of strategic actions between platform owners and complementors. This study aims to explore how platform owners and complementors drive the evolution of platform ecosystems through strategic synergy implementation, driven by market triggers, and the outcomes of these synergistic strategies.Design/methodology/approachThe study adopts a longitudinal case study of the Xiaomi platform ecosystem. The temporal bracket and structural approach are used to analyze and theorize the data.FindingsThis study examines the strategic interactions between platform owners and complementors and identifies three synergistic strategies, namely, resource synergy strategy, system synergy strategy and scenario synergy strategy. These strategies build on the foundation of core products, facilitating horizontal diversification and integration while driving vertical contextualization. By implementing these synergistic strategies, platform ecosystems evolve progressively, starting with one based on core products, then moving to one based on combination products and finally to one based on scenario products.Research limitations/implicationsFurther research can explore the distinctions in the evolutionary trajectories and synergistic strategies of transactional and innovative platform ecosystems. Additionally, by delineating the types of complementors, studies can investigate the strategic interactions between different types of complementors and platform owners, as well as the synergistic strategies, providing new perspectives and deeper insights.Originality/valueThis study expands the procedural research on the evolution of platform ecosystems, revealing the evolutionary trajectory of platform ecosystems. Meanwhile, we also pay attention to the correlation of strategic actions between platform owners and complementors, proposing three synergistic strategies to drive the evolution of platform ecosystems.

Lu Sun, Hui He, Wenmin Lin

A major challenge for managers in the digital economy era is reduce uncertainty. Information interaction capabilities (IICs) have been proposed as an organizational capability that facilitates the creation of value by leveraging uncertainty. This is achieved by assisting managers in the configuration, application, and integration of diverse information interaction resources into novel capabilities within an emerging digital intelligence technology environment. However, due to the elusive nature of IICs, managers may encounter difficulties in creating value in digital economy if they are unable to accurately understand the formation mechanism of IICs. Therefore, we first seek to propose the components of IICs. Specifically, by the Gioia methodology, we identify a set of capabilities of IICs—bianyi capability (triadic mutability capability), self-cognizance capability (introspective capability), contextual empathy capability, and ecosystem orchestration capability. Second, through a longitudinal case analysis of Xiaomi’s evolutionary trajectory—segmented into three distinct phases based on milestone events—we systematically studied the formation mechanism of IICs. Xiaomi has demonstrated a relatively completely and weak IICs since its inception. Through case studies of its evolutionary trajectory, we observe that IICs has progressively evolved and strengthened via four interconnected dimensions, forming a mature and robust IICs. Finally, we explore the impact of IICs on competitive advantages and construct a framework of competitive advantage suitable for an environment of digital economy. The study’s implications for creating value in digital economy environments by revealing components and formation mechanism of IICs are discussed.

Qiufen Wang, Xue Fan, Ziqiang Fu, Jiahao Wang

ABSTRACT This study investigates consumer perceptions of Xiaomi’s brand extension from electronics to automobiles, focusing on the SU7 electric car. Guided by signaling theory, it explores how brand trust, perceived quality, and user satisfaction influence purchase intention. A cross-sectional survey of 200 Chinese consumers and deep learning analysis of 12,843 social media comments were conducted. BERT was used for sentiment analysis, LDA for topic modeling, and co-occurrence analysis to uncover potential purchase intentions. Regression results showed brand trust and satisfaction with Xiaomi’s phones were the strongest predictors of purchase intention (R2 = 0.868, p < 0.01), while innovation played a secondary role. Sentiment analysis revealed generally positive attitudes toward value for money, technology, and brand loyalty, with some concerns about quality control. This study extends signaling theory by demonstrating how brand equity can transfer to high-involvement products and offers strategic insights for tech firms expanding into trust-sensitive industries.