Core Viewpoint - Li Auto officially announced its entry into the markets of Egypt, Kazakhstan, and Azerbaijan, following its expansion into Uzbekistan, launching three main models: Li L9, Li L7, and Li L6, while providing official warranty services, professional after-sales support, and ongoing OTA technology upgrades for overseas users [1] Group 1 - Li Auto has established research and development centers in Germany and the United States [1] - The new product set to launch in 2026 will consider overseas market regulatory compliance from the initial development stage [1] - This strategy aims to enhance the company's competitiveness in the global market [1]

Core Viewpoint - Li Auto has reported a significant increase in organized attacks and defamation against the company and its vehicle owners, including personal information breaches and false information dissemination [1] Group 1: Company Actions - The legal department of Li Auto has stated that it will use legal means to protect its brand and user reputation, aiming to maintain a clear online environment and fair market competition [1] - Li Auto has highlighted that the organized and premeditated attacks are typical criminal activities of "black water armies" [1] Group 2: Law Enforcement Response - The public security bureau in Yantai, Shandong Province, has successfully uncovered and dismantled a criminal organization involved in these online attacks, arresting 12 suspects and seizing over 1 million yuan in illicit funds [1] - The operation led to the shutdown of more than 8,000 illegal online accounts, demonstrating a comprehensive crackdown on this new type of "internet water army" criminal group [1]

Group 1 - Xiaomi officially announced the open-source release of Xiaomi MiMo-V2-Flash, a MoE model with a total parameter count of 309 billion (15 billion activated), achieving top 2 in global open-source model benchmarks [1] - The model features innovations such as Hybrid attention architecture and multi-layer MTP inference acceleration, resulting in a code capability comparable to the closed-source model Claude 4.5 Sonnet, but at only 2.5% of its inference cost and with a 2x increase in generation speed [1] - Xiaomi MiMo-V2-Flash outperformed DeepSeek V3.2 and K2-Thinking in most evaluation benchmarks, reducing parameter count by 50% to 67%, and achieving low cost and high speed, with preliminary capabilities to simulate the world [1] Group 2 - The next generation of intelligent agent systems is envisioned not merely as "language simulators" but as true "intelligent agents" that understand and coexist with the human world [2] - There is a shift in agent execution capabilities from merely "answering questions" to "completing tasks," incorporating memory, reasoning, autonomous planning, decision-making, and execution abilities [2] - Unified multimodal perception is essential for understanding the physical world, which will enhance integration with smart devices like glasses [2]

Core Insights - Tencent has upgraded its large model research architecture by establishing new departments, indicating a strategic focus on enhancing its AI capabilities [1][2] - The recruitment of top AI talent, including Yao Shunyu from OpenAI, signifies Tencent's commitment to strengthening its position in the global AI competition [1][3] Group 1: Organizational Changes - Tencent has formed the AI Infra, AI Data, and Data Computing Platform departments to bolster its large model research and core capabilities [1][2] - Yao Shunyu has been appointed as the Chief AI Scientist and will oversee the AI Infra and large language model departments [1][2] - The AI Infra department will focus on distributed training and high-performance inference services, while the AI Data and Data Computing Platform departments will handle data and evaluation systems [2] Group 2: Talent Acquisition - Tencent is actively recruiting top AI researchers, including poaching talent from ByteDance with offers of double salaries [3] - Yao Shunyu's recruitment has already led to the acquisition of additional talent from various AI firms [3] - The company is offering salaries 50% above industry standards to attract fresh PhD graduates [3] Group 3: Competitive Landscape - The ongoing recruitment efforts reflect the intense global competition for AI talent, with Tencent aiming to solidify its standing in the industry [3] - Despite previous advancements, Tencent is still in the exploratory phase regarding model development, indicating room for growth [3]

凤凰网科技讯12月17日，多益网络昨日发布声明：近日，美国《华尔街日报》发布关于多益网络董事长 徐波先生的恶意造谣报道(题为"The Chinese Billionaires Having Dozens of U.S.-Born Babies Via Surrogate")，称"在社交媒体上，(徐波)他的公司表示，他通过美国代孕拥有超过100个孩子"。该报道蓄 意混淆事实，捏造不实信息，对多益网络董事长徐波先生的个人声誉及多益网络形象造成严重侵害。 在此，我公司严正要求《华尔街日报》立即删除相关造谣报道，就其恶意歪曲事实的行为向多益网络董 事长徐波先生公开道歉，并通过其全球发行渠道发布澄清声明，消除不良影响。 请转发虚假谣言的各媒体、自媒体第一时间自行删除相关谣言，针对此次造谣事件对多益网络董事长徐 波先生造成的名誉损害，我公司已委托律师团队进行证据固定，对于相关造谣、传谣的各平台用户，多 益网络将保留追究法律责任的权利。 在此，多益网络正告《华尔街日报》及相关造谣、传谣的各平台用户，停止对多益网络董事长徐波的恶 意诋毁，多益网络强烈谴责并坚决反对此类捏造事实的造谣、传谣行为。 对于《华尔街日报》的不实报道， ...

Group 1 - The head of material technology at Li Auto, Ji Chaochao, responded to concerns about the repairability of integrated die-casting structures, stating that these structures are designed with multiple layers of protection during collisions [1] - In low-speed collisions, typically only the crash beam may be damaged, which can be easily replaced, while higher energy collisions may damage the energy-absorbing box and longitudinal beams, which can also be replaced as a whole [1] - The design of integrated die-casting structures aims to efficiently absorb energy within an appropriate deformation range, and the claim that minor collisions would necessitate a complete vehicle replacement is unfounded [1] Group 2 - International car manufacturers, such as Volvo with its EX90 model, are utilizing integrated die-casting technology, indicating a growing acceptance of this method in the industry [2] - The primary reason for not using integrated die-casting in many domestic vehicles is cost rather than repairability, with lower-cost vehicles typically opting for steel bodies due to their affordability [2] - Some domestic brands in the price range of 200,000 to 400,000 yuan also avoid integrated die-casting, which reflects a "hidden cost advantage" in their manufacturing choices [2]

凤凰网科技讯12月17日，乐道汽车官宣发布L60紫罗兰版，该版本具有紫白双拼限定配色，限量666 台，12月18日上市即交付。 12月5日，乐道汽车发布了乐道L60黑骑士版，同样限量发售666台，整车购买21.19万，租电方式购买 15.49万，吴尊是首位用户。该版本拥有星空黑车色、20寸星芒运动熏黑轮毂，标配曜黑品牌标志及字 标；内饰采用雅黑高级皮革座椅，并配以橙色内饰缝线及安全带作为点缀；标配后排右侧折叠桌板。 据了解，普通版L60后排标配座椅通风，底盘舒适度提升，CLTC百公里能耗降到了11.9kWh，新增了20 英寸轮圈、后排右侧折叠小桌板及地图袋，整车购买20.69万元起，租电14.99万元起。 ...

据介绍，WholeBodyVLA通过从人类第一视角视频中学习移动与操作的潜在动作表示，以降低对昂贵机 器人遥操作数据的依赖。同时，研究团队设计了一种面向移动操作的强化学习控制器，通过简化控制目 标来提升运动执行的稳定性。 实验验证显示，该框架使机器人能在不同起始位置走向目标并完成操作，并对场景和物体表现出一部分 泛化能力，在干扰地形上也能保持基本的移动方向与平衡。研究团队表示，该工作为将视觉-语言-动作 （VLA）范式扩展至双足人形机器人全身控制提供了可行路径。 凤凰网科技讯 12月17日，让人形机器人能够像人一样完成装箱、搬运、推车等移动操作任务，一直是 人们对具身智能的期待。近日，来自香港大学、智元AGIBOT、复旦大学和上海创智学院的联合研究团 队提出了WholeBodyVLA，一种面向真实世界的人形机器人全身 Vision–Language–Action框架。该工作 基于智元灵犀X2研究发布，将VLA扩展至双足人形机器人的全身控制，验证了其在全身loco- manipulation任务中的可行性。 与原地操作相比，loco-manipulation的难点不在于单一技能，而在于行走与操作必须在同一任务中 ...

Core Viewpoint - SenseTime has launched the "Yuying Marketing Intelligent Body," an AI solution aimed at automating e-commerce operations, including store management, live streaming, and traffic allocation [1] Group 1: Product Features - The intelligent system consists of five functional units: Store Operation Agent, Live Streaming Operation Agent, Traffic Allocation Agent, Live Streaming Control Agent, and Digital Human Agent [1] - These agents can automatically perform tasks such as data organization, review analysis, monitoring of ad placements, and live streaming assistance [1] Group 2: Efficiency Improvements - In certain scenarios, the efficiency of store operation tasks can increase by up to 20 times, and the time required for live streaming reviews can be reduced from 2 hours to 20 minutes [1] Group 3: Unique Selling Proposition - The Yuying Digital Human Agent can replicate real human live streaming without complex video capture, enabling "same-day replication for same-day broadcasting," addressing the challenges faced by small and medium-sized businesses regarding the limited availability of human hosts [1] Group 4: Technical Foundation - The system operates on SenseTime's "Riri New" large model and integrates with an intelligent agent framework capable of operational tasks [1] - The system has been applied in real commercial scenarios over the past year and has been adapted to work with major domestic hardware chips [1]

Core Viewpoint - Muxi Co., Ltd., known as the "second domestic GPU stock," officially listed on the Sci-Tech Innovation Board, with a significant first-day stock price increase of 692.95%, reflecting strong market interest in domestic high-end GPU chips [1] Company Overview - Muxi Co., Ltd. was founded in 2020 by former AMD expert Chen Weiliang and has successfully mass-produced its first AI inference chip in 2023 [1] - The company plans to launch the Xiyun C600 chip in 2025, aiming to achieve a complete domestic supply chain from design to packaging and testing [1] - The next-generation Xiyun C700 series is scheduled for tape-out in the second half of 2026, with performance targeting industry-leading products [1] Industry Context - The listing of Muxi Co., Ltd. signifies that leading domestic GPU companies have gathered on the Sci-Tech Innovation Board, following the example of Moore Threads [1] - The demand for AI computing power is surging, driven by technological self-sufficiency strategies, as domestic GPU companies accelerate their transition from "usable" to "highly usable" products despite fierce competition from international giants and domestic peers [1]