2026年，人工智能将成为技术与工业界中最亮眼的明星。Granter预测，2026年全球人工智能（AI）总支出将达到2.52万亿美元，同比增长44%。作为科技界 代表的IBM公司，宣布在2025年第四季度已经实现生成式人工智能业务规模突破125亿美元，公司在营收、利润和自由现金流方面的表现均超预期。 所谓因果AI或称为因果推理的出现，是因为当前的AI智能决策体系主要基于由大语言模型LLM所构建的知识图谱，而智能体在进行决策的过程中还需要理 解决策与结果之间的因果关系，也就是事件为什么、如何发生，以及建立与此相应的数学模型。 在2022年的Gartner人工智能技术成熟度曲线中，因果AI处于技术萌芽期，到了2026年开始进入到实用阶段。IBM专家指出，因果推理主要为决定一个变量是 否导致另一个变量变化的过程。因果推理算法源于流行病学、公共健康、计算经济学和数据科学等不同领域，并广泛在健康医疗、社会科学和各类决策中应 用。 在CES 2026上有很多机器人等令人眼花缭乱的产品，但总结下来体现了智能终端从设备迈向智能生产生活方式的关键转折，无论是企业用户还是消费者的 关注点都从单一硬件转向产品能否切实节省时间、降低 ...

Recent Events - The founder of Shengda Group, Chen Tianqiao, released an internal letter on February 6, 2026, outlining the technological direction of MiroMind, focusing on "discovery-based intelligence" and "general solvers," while avoiding the general chatbot arena [1] - The emphasis is on promoting AI development through causal reasoning and systematic innovation, which may indirectly impact Shengda Technology's business ecosystem [1] Stock Recent Trends - Historical data indicates that Shengda Technology's stock price experienced multiple fluctuations in December 2025, including a surge of 5.34% on December 29 and a drop of 5.00% on December 17 [2]

Core Viewpoint - The article emphasizes the need for robots to possess a "thinking" brain that understands the causal relationships of the world, rather than merely focusing on perfecting their physical forms. This perspective is articulated by Bai Huiyuan, the founder and CEO of Infiforce, who argues that the essence of embodied intelligence lies in the brain's ability to perceive and predict the physical world [1][2][3]. Group 1: Industry Context - In 2023, Bai Huiyuan left Alibaba to establish Infiforce amidst a competitive landscape where many companies were focused on hardware advancements, leading to a "body-making" arms race in the robotics industry [2]. - The robotics industry is currently characterized by a fascination with hardware, with companies competing on joint flexibility and human-like movements, while neglecting the cognitive capabilities of the robots [2][3]. - Infiforce aims to break this trend by focusing on developing a "thinking" brain that can adapt to various bodies and understand the physical world, rather than merely enhancing hardware specifications [3][12]. Group 2: Technological Approach - Infiforce's technological strategy involves a continuous learning model called Hyper-VLA combined with a causal world model, which contrasts with the mainstream AI approach that primarily relies on correlation [5][6]. - The existing AI models often depend on vast amounts of data for training, which is not feasible in the physical world, leading to issues of data scarcity and lack of robustness [6]. - Infiforce's approach integrates causal reasoning into its models, allowing robots to understand the implications of their actions, thus enhancing their decision-making capabilities in unfamiliar environments [6]. Group 3: Business Development - In 2025, Infiforce secured over 500 million yuan in commercial orders, signaling a significant milestone in the industry, although these orders are seen more as experimental partnerships rather than the launch of standardized products [8]. - The orders came from leading clients in various sectors, including cultural tourism, research, energy, and smart manufacturing, indicating a willingness to invest in the potential of robotics beyond mere demonstrations [8]. - Infiforce's AstroDroid AD series is transitioning from demonstration to pilot projects, where robots are actively engaging in real-world tasks, such as understanding visitor intentions in museums and performing household chores [8]. Group 4: Vision and Future Aspirations - Bai Huiyuan envisions Infiforce becoming an integral part of the robotics ecosystem, akin to "air" and "water," where the core intelligence of future robots will stem from Infiforce [13]. - The ultimate goal is to create robots that seamlessly integrate into human environments, making their intelligence so advanced that users forget they are interacting with machines [13].

Core Insights - The article discusses the advancements in AI models for gaming, particularly focusing on the Pixel2Play (P2P) model developed by researchers at Player2, which aims to enhance AI's performance in real-time gaming environments [2][5]. Group 1: Model Development - The P2P model utilizes game visuals and text instructions as inputs to generate corresponding keyboard and mouse operation signals, achieving over 20Hz end-to-end inference speed on consumer-grade RTX 5090 graphics cards [2]. - P2P has been trained on over 40 games, totaling more than 8300 hours of gameplay data, and can play multiple games on Roblox and Steam in a zero-shot manner [2]. - The model employs a lightweight framework and is built from scratch, featuring a decoder Transformer and a lightweight action-decoder to enhance inference speed by five times [10]. Group 2: Training Data and Open Source - High-quality "visual-action" data is scarce online, prompting the Open-P2P project to open-source all training datasets to fill this gap [5][3]. - The training data includes game images, text instructions, and precise keyboard and mouse operation annotations, which are crucial for training effective game AI models [8][5]. Group 3: Model Evaluation - P2P has been evaluated using four different model sizes, with parameters ranging from 150M to 1.2B, achieving inference speeds of 80Hz for the 150M model and 40Hz for the 1.2B model [12]. - In human evaluations, the 1.2B model showed a preference rate of 80%, 83%, and 75% over smaller models in various games, indicating superior performance [13]. - The model's ability to follow text instructions significantly improved its success rate in tasks, demonstrating strong understanding and execution capabilities [15]. Group 4: Causal Reasoning - The article highlights the challenge of causal confusion in behavior cloning, particularly in high-frequency interaction environments, and notes that increasing model size and training data can enhance the model's understanding of causal relationships [17]. - As training data and model parameters increase, the P2P model's performance in causal inference assessments shows a positive trend [19].

Group 1 - The core viewpoint of the news highlights the significant market interest in intelligent driving technologies, particularly focusing on companies like索菱股份, which has shown a 10% increase in stock price due to its advancements in the field [1][3] -索菱股份 has accumulated substantial experience in research and delivery in the field of assisted driving, with successful mass production cases and driving data in L2 and L2++ products, and feasible technical solutions for L3 autonomous driving [1] - The company serves a diverse customer base that includes both passenger and commercial vehicles, covering various models such as sedans, SUVs, MPVs, and light trucks [1] Group 2 - NVIDIA has launched the Alpamayo series of open-source AI models, simulation tools, and datasets aimed at enhancing the development of safe and reliable inference-based assisted driving vehicles [2] - The Alpamayo initiative introduces a complete open-source closed loop that incorporates "causal reasoning" capabilities into the autonomous driving research paradigm, significantly lowering the barriers to industry implementation [2] - The open-source framework includes a high-fidelity end-to-end simulation environment and a diverse large-scale dataset with over 1,700 hours of driving data, which supports rapid validation and strategy optimization for autonomous driving [2]

Core Viewpoint - The article discusses the challenges faced by traditional single-agent systems in long-cycle, multi-step collaborative tasks, highlighting the need for a distributed agent framework with global planning and causal dependency management capabilities [1][2]. Group 1: CausalMACE Method - The CausalMACE method is proposed by a research team from Hong Kong University of Science and Technology and Tencent, integrating causal reasoning mechanisms into open-world multi-agent systems to provide scalable engineering solutions for complex task collaboration [2]. - The method includes a "global causal task graph" concept, allowing AI to learn "if-then" logic, enabling dynamic adjustments and clear division of labor among agents [5][6]. Group 2: Framework Components - The CausalMACE framework consists of three main components: Judger, Planner, and Worker [7]. - Judger ("裁判") verifies the legality of actions in real-time and provides feedback on success or failure, ensuring all agents operate under the same game rules [11]. - Planner ("总工") breaks down complex tasks into smaller sub-tasks and creates a rough flowchart based on game rules, refining it through causal reasoning to ensure task dependencies remain valid [12][14]. - Worker ("调度室") utilizes depth-first search to split the causal graph into multiple production lines, calculating a "busy index" for real-time task reassignment among agents [16]. Group 3: Experimental Results - The experimental results indicate that CausalMACE significantly enhances both completion rates and efficiency in benchmark tasks such as construction, cooking, and escape rooms, achieving up to a 12% increase in task completion rates and a maximum efficiency improvement of 1.5 times compared to baseline methods [17]. - In the VillagerBench benchmark tasks, CausalMACE outperformed AgentVerse and VillagerAgent across various metrics, demonstrating its effectiveness in multi-agent collaboration [18]. Group 4: Author Information - The lead author of the paper is Professor Wang Hao, an assistant professor and doctoral supervisor at Hong Kong University of Science and Technology (Guangzhou), with a research background in generative AI models and 3D reconstruction [19][20].

Core Viewpoint - The article discusses recent advancements in the field of embodied intelligence, particularly focusing on the integration of large language models (LLMs) with reinforcement learning and world models, highlighting several notable research papers from 2024 [2][3]. Group 1: UniSim - UniSim aims to learn general real-world interactive simulators through generative modeling, revealing that natural datasets can provide diverse advantages for learning simulators [3]. - The research demonstrates that integrating various datasets allows for the simulation of high-level commands and low-level controls, enabling zero-shot application in real-world scenarios [3]. Group 2: Robust Agents - The study from Google DeepMind asserts that causal reasoning is essential for robust and general AI, concluding that agents capable of satisfying regret bounds must learn approximate causal models [5]. - This finding has significant implications for transfer learning and causal inference [5]. Group 3: MAMBA - MAMBA introduces an efficient world model approach for meta-reinforcement learning, addressing sample efficiency issues prevalent in current methods [8]. - The framework shows a remarkable improvement in sample efficiency, achieving up to 15 times better performance in high-dimensional tasks [8]. Group 4: EMMA - EMMA leverages LLMs trained in text-based worlds to guide the training of visual world agents, enhancing their ability to interact with dynamic environments [10]. - The approach results in a significant success rate improvement of 20%-70% in diverse tasks compared to existing VLM agents [10]. Group 5: Text2Reward - The Text2Reward framework automates the generation of dense reward functions using LLMs, addressing the challenges of reward function design in reinforcement learning [13][14]. - The method demonstrates superior performance in 13 out of 17 tasks, achieving over 94% success in new motion behaviors [14]. Group 6: Online Continual Learning - The research proposes two frameworks for continuous learning in interactive instruction-following agents, emphasizing the need for agents to learn incrementally as they explore their environments [17][18]. - A confidence-aware moving average mechanism is introduced to update parameters without relying on task boundary information [18]. Group 7: AMAGO - AMAGO is a scalable contextual reinforcement learning framework that addresses challenges in generalization, long-term memory, and meta-learning [21]. - The framework allows for parallel training of long-sequence transformers, enhancing scalability and performance in complex tasks [21]. Group 8: PDDL-based Planning - The study presents a novel paradigm for task planning using pre-trained LLMs, focusing on building explicit world models through PDDL [22][23]. - The framework significantly reduces the need for human intervention by allowing LLMs to convert between PDDL and natural language, facilitating efficient model correction [23].

Group 1 - The article discusses whether LLMs (Large Language Models) truly "understand the world" or if their reasoning is merely a form of pattern matching, highlighting the debate within the industry regarding the nature of LLM reasoning capabilities [1][3][4] - It references a paper from Apple suggesting that current reasoning models do not genuinely think, but rather engage in pattern matching, which has sparked renewed discussion in the AI community [3][4] - The article mentions that true reasoning involves understanding causal relationships, as emphasized by various researchers, indicating that LLMs lack a causal framework necessary for deep and flexible reasoning [5][6][7] Group 2 - The article explores the motivations behind enterprises increasing their spending on generative AI, noting a shift from building in-house solutions to purchasing third-party AI applications [1][2] - It outlines the evaluation framework for selecting AI models, which includes key factors that influence procurement decisions in the context of traditional software purchasing characteristics [1][2]