Core Viewpoint - The article discusses the development of OpenScholar, an AI assistant designed specifically for researchers to synthesize scientific literature accurately and efficiently, addressing the issue of "hallucination" in existing large language models [2][5][21]. Group 1: OpenScholar Overview - OpenScholar is a retrieval-augmented language model that can intelligently retrieve relevant paragraphs from 45 million open-access papers and generate comprehensive review papers with accurate citations [5][7]. - The model's citation accuracy is comparable to that of human experts and surpasses mainstream models like GPT-4o in multiple tests [5][11]. Group 2: Functionality and Workflow - OpenScholar operates through a three-step process: retrieval of relevant content, generation of answers with citations, and self-feedback for iterative improvement [7][9]. - The system is built on a dedicated data store (OpenScholar DataStore) that allows for transparent and reproducible research [7][21]. Group 3: Evaluation and Performance - The ScholarQABench benchmark was developed to assess AI systems' reliability in synthesizing scientific literature, featuring nearly 3,000 expert-written questions across various fields [12][13]. - OpenScholar demonstrated impressive results in the benchmark, outperforming GPT-4o in citation accuracy and overall usefulness, with human experts favoring OpenScholar's responses over those of GPT-4o [16][18][19]. Group 4: Implications for Research - The introduction of OpenScholar signifies a significant advancement in the application of AI in scientific research, potentially transforming literature reviews from a burdensome task into an efficient exploration process [21][23]. - Future developments may enhance OpenScholar's capabilities, making it a true collaborator for researchers, allowing them to focus more on innovation rather than information filtering [23].

回顾人类历史，科学逐渐演变为经济增长的主要驱动力，重塑了20世纪的经济结构和周期，直接影响着 宏观经济格局和微观经济模式。正如智能手机带来的划时代影响所展现的，科技创新不仅成为经济增长 的重要引擎，也从根本上改变了我们每个人的日常生活方式。然而，我们也必须认识到，人类正处于一 个充满前所未有、相互交织挑战的时代。科学在不断取得重大突破的同时，也积累了一定的系统性、结 构性和制度性风险。人工智能正对科技产生全面且加速的影响，从根本上重塑科学生态系统。人类正站 在选择未来科学发展方向的十字路口。 科学资源本质上是人类集体智慧的结晶。在科学真理面前，人人生而平等。每个人都应该能够获取、公 开使用和再利用科学知识，从而实现科学领域的公平正义。本次峰会的主要目标，就是推动形成开放科 学共识，使科学成为全人类共同的事业。开放科学旨在超越自然、人工智能、世代、国界、价值体系和 文明的界限，促进科学知识的自由流动和广泛共享，构建适应数字化、智能化时代的新型科学框架。 我们相信，从现在到2050年，人工智能发展的技术奇点将正式到来。现代科学的使命是确保这一进程可 以推动人类文明进步。科学的使命不仅在于解释世界，更在于推动文明迈 ...

当今世界，主要国家的科技发展战略不仅影响本国，也影响全球。中国在科技发展战略和政策方面的选 择，日益受到全球科学界的瞩目 2月1日至3日，世界顶尖科学家协会将在阿联酋迪拜举办世界顶尖科学家峰会。届时，来自多个国家的 71位顶尖科学家、26位一流大学校长和医院院长，以及22位青年科学家将齐聚一堂，围绕"基础科学： 应对人类未来的挑战"这一主题，展开深入而开放的讨论。 2017年11月，我与迈克尔·莱维特、巴瑞·夏普莱斯等多位诺贝尔奖得主在中国香港成立世界顶尖科学家 协会。协会成立次年，在中国政府的大力支持和推动下，首届世界顶尖科学家论坛在上海开幕，成为科 学界的年度盛会。习近平主席分别向第二届和第三届世界顶尖科学家论坛致贺信、作视频致辞，充分体 现了中国对推进前沿科技发展和促进国际科学合作的高度重视。 当今世界，主要国家的科技发展战略不仅影响本国，也影响全球。中国在科技发展战略和政策方面的选 择，日益受到全球科学界的瞩目。近年来，中国秉持开放合作理念，在基础研究领域取得显著进展，尤 其是在人工智能、材料科学、生命科学和新能源等领域。凭借市场化的创新机制、雄厚的人才储备和广 阔的应用环境，中国在推动产业创新方面取 ...

Core Viewpoint - Open science is increasingly recognized as an effective means to promote the sharing, communication, and collaboration of scientific knowledge, with major global innovation centers, including Beijing, placing significant emphasis on it [1][2]. Group 1: Achievements in Open Science - Beijing has seen a rise in the proportion of open access research papers from 26.1% in 2015 to 41.4% in 2024 [2]. - The establishment of the Beijing International Big Data Exchange has led to the creation of several scientific data centers, facilitating the reuse of scientific data [2]. - Collaborative efforts have resulted in the integration of 160 million academic papers into large academic models, showcasing initial success in AI data reuse [2]. Group 2: Areas for Improvement - Current regulations lack unified, clear, and specific guidance on optimizing research outcomes, infrastructure, and international cooperation [3]. - There is a need for standardized protocols regarding the structure, content, and quality of scientific data, as well as improved interoperability among different data centers and platforms [3]. - Recommendations include establishing a collaborative mechanism among institutions and creating a "Beijing Open Science Cloud Platform" to integrate key knowledge resources [3]. Group 3: Future Goals and Strategies - The goal is to increase the proportion of open access papers to 60%-80% by 2030, aligning with levels seen in developed countries [4]. - Emphasis will be placed on strengthening scientific data collection and exploring mandatory data submission systems for publicly funded research [4]. - The establishment of an international open science fund is proposed to support developing countries in utilizing Beijing's scientific facilities [5].

Core Viewpoint - The launch of the domestic preprint platform "Langtaosha" marks a significant step for China's life sciences sector, enabling independent and secure management of research data and academic dissemination, while addressing the challenges posed by international publication systems [1][2]. Group 1: Platform Overview - "Langtaosha" is a preprint platform developed collaboratively by Shenzhen Medical Academy, Shenzhen Bay Laboratory, Tsinghua University, and Westlake University, aimed at enhancing the efficiency and safety of academic communication in China [1][2]. - The platform allows researchers to publish their findings without incurring publication fees, contrasting with the traditional Open Access (OA) model that often requires authors to pay article processing charges (APC) [2][3]. Group 2: Market Context - The global academic publishing market is predominantly controlled by developed countries, with major players like Elsevier and Springer dominating the landscape, leading to over 90% of SCIE papers from Chinese scientists being published in foreign journals [2][3]. - In 2023, Chinese authors paid a total of $750 million in APCs, reflecting a 23% annual growth rate, which highlights the financial burden on researchers and the need for alternative publishing avenues [3]. Group 3: Benefits of Preprint Platforms - Preprint platforms like "Langtaosha" facilitate rapid dissemination of research findings, allowing for immediate feedback and collaboration across disciplines, thus enhancing the overall quality and impact of research [4][5]. - The introduction of blockchain technology in "Langtaosha" ensures transparency and traceability of research records, which is crucial for maintaining academic integrity and prioritizing original contributions [4]. Group 4: Future Development and Recommendations - Experts suggest that the development of preprint platforms in China is still in its exploratory phase and requires improved top-level design, institutional norms, and academic evaluation systems to support high-level scientific independence [6][7]. - Recommendations include establishing a national certification system for preprint platforms, defining qualification criteria, and integrating preprints into existing research management frameworks to enhance their credibility and acceptance [6][7].

Group 1 - The World Leading Scientists Summit (WLS) and the World Government Summit (WGS) will hold a joint summit in Dubai from February 1 to 3, featuring 71 top scientists, including 39 Nobel laureates [1][2] - The theme of the summit is "Fundamental Science: Addressing the Challenges of Humanity's Future," focusing on four key topics: the integration of artificial intelligence and fundamental science, the co-evolution of scientific progress and social development, global collaboration in open science, and the future of scientific cooperation systems [1] - The summit aims to enhance the UAE's position as a "global scientific hub" through extensive scientific collaboration [1] Group 2 - The World Government Summit, held annually in Dubai since 2013, aims to drive global governance transformation through innovation and technology, with over 6,000 participants last year and an expected attendance of 40 heads of state and over 500 government ministers this year [2] - The joint summit model is expected to create new opportunities for resource linkage and collaboration, enhancing the depth and breadth of engagement between the two events [2]

场。大会组委会 供图 主旨报告环节，中国工程院院士、华南理工大学教授陈克复，剑桥大学教授Ali Reza Kamali，马来西亚 理工大学副校长Ali Selamat等，围绕科技重塑产业竞争力与可持续发展作专题学术报告。 大会现 此外，同期举办的JOFM期刊分会、"学术生态重构"研讨会及多场专题论坛，汇聚国际出版协会、全球 知名期刊专家，探讨开放科学与人工智能浪潮下，如何构建更开放、智能、可信的新型学术生态。会议 议题覆盖新能源、人工智能、电子信息等重点领域，AiScholar艾思科蓝以"人工智能+科研学术"服务模 式，赋能跨学科学术对话与协同创新。(完) 大会由中国科学报社、华南理工大学等单位联合主办，以"汇聚国际科研创新智慧、共谋学术生态高质 发展"为主题，采用"1+6+N"立体化会议体系，涵盖主论坛、国际期刊发展论坛、多场专题研讨会及"研- 产-教"一体化成果展。 大会上，华南理工大学计算机科学与工程学院与马来西亚理工大学计算机学院签署合作协议，共建国际 科研交流平台。《中国国际学术会议发展蓝皮书2025》发布，该蓝皮书由广州市科协指导，为学术会议 质量提升与国际化发展提供参考。 中新网广州12月21日 ...

Core Insights - The French National Centre for Scientific Research (CNRS) announced it will stop accessing the Web of Science database from January 1, 2026, as part of its commitment to open science and changing the evaluation of scientific work [1][7][9] Group 1: Open Science Policy - CNRS has been promoting open science since the release of its first Open Science Roadmap in 2019, opposing quantitative evaluations based on bibliometric indicators in favor of qualitative assessments [3][9] - The decision to discontinue Web of Science is a significant step in CNRS's open science policy, aiming to enhance the quality and ethics of scientific evaluation [4][10] Group 2: Financial Implications - By stopping the subscription to Web of Science, CNRS will save €1.4 million annually, which will be reinvested into open science initiatives, particularly the development of open databases [5][11] Group 3: Transition to OpenAlex - CNRS is encouraging researchers to transition to OpenAlex, which offers greater visibility for non-English journals and a broader range of indexed journals compared to Web of Science [5][11] - OpenAlex aims to address the coverage gaps in various fields, such as humanities, social sciences, computer science, and mathematics, where Web of Science has been lacking [5][11] Group 4: Challenges and Future Directions - While leaving Web of Science may temporarily hinder the ability to make standardized comparisons between institutions and researchers, CNRS acknowledges the need to establish new principles for such evaluations [6][12] - CNRS is actively working to improve OpenAlex's metadata management to enhance its potential and is confident that it will soon meet the needs of researchers [6][12] - The organization aims to regain autonomy in scientific decision-making and enhance transparency through open tools, inviting partners to also distance themselves from Web of Science [6][12]

Core Insights - The article highlights the significant growth of Chinese researchers' publications in Cell Press journals over the past decade, showcasing advancements in originality, systematic approaches, and openness in Chinese scientific research [2][48]. - In 2024, the total number of research papers published by Chinese institutions in Cell Press journals reached a record high of 2,446, marking a 17.6% increase from the previous year and nearly quadrupling the output from 2020 [3][14]. Publication Growth - The number of papers published in the life sciences field by Chinese scholars was the highest in 2024, totaling 860 papers, with the medical field showing the most notable growth, increasing by 138 papers [6]. - The journal Cell Reports Medicine saw the number of Chinese papers surpass 100 for the first time, while materials science and interdisciplinary fields also experienced steady growth, reaching 554 and 394 papers, respectively [6]. Journal Performance - In 2024, 18 hybrid journals under Cell Press reported growth in the number of Chinese publications, with Joule and Device in materials science showing the highest increases of 32 and 31 papers, respectively [7]. - The medical journal Med saw a remarkable increase, with a total of 32 papers published, reflecting a 2.5-fold increase from the previous year [7]. Open Access Trends - The proportion of papers published in open access (OA) format by Chinese scientists in Cell Press journals approached three-quarters of all Chinese publications, totaling 1,761 papers, which is a 13% increase from the previous year [10]. - The number of OA papers published in hybrid journals rose from 69 to 119, marking a 72.5% increase, indicating a growing commitment to open access among Chinese researchers [11]. Annual Paper Awards - The 2024 Cell Press Annual Paper Awards recognized 50 outstanding papers across five fields, with an additional 25 papers voted as the most popular by the public [14]. Research Institution Landscape - In 2024, the number of Chinese institutions publishing in Cell Press journals exceeded 300 for the first time, with a notable increase of 40 institutions from the previous year [32]. - Sun Yat-sen University led in publication volume with 113 papers, followed by Shanghai Jiao Tong University and Zhejiang University with 106 and 105 papers, respectively [32]. Regional and International Collaboration - Beijing remained the leading region for publications with 442 papers, followed by Guangdong and Shanghai [38]. - The number of countries and regions collaborating with Chinese scholars reached 76, reflecting a 20.6% increase, with the United States being the most frequent collaborator [43][47].

Core Insights - The Fourth World Frontier Technology Conference was held with the theme "Technology Creates a Better Future," focusing on global technology resource integration and innovation value realization [1][7] - The conference featured discussions on cutting-edge technologies such as artificial intelligence, metaverse applications, immune cell therapy, and mechanical manufacturing, addressing the challenges of transforming research into industry [1][2] Group 1: Conference Highlights - The conference included a rich agenda with an opening ceremony, high-level dialogues among academicians, exhibitions of frontier technology achievements, and international project cooperation meetings [2] - A notable report by Chinese Academy of Sciences Academician Chen Runsheng discussed breakthroughs in AI applications in fields like healthcare and materials [2] - The Secretary-General of the China High Technology Industry Development Promotion Association emphasized the need for a dual approach of "technological breakthroughs + intellectual property protection" for high-quality innovation [2] Group 2: Technological Innovations - The conference introduced the "Cloud Network," integrating metaverse exhibition halls, AR interactive technology, smart meeting systems, MR mixed reality technology, and blockchain, enabling global participation and establishing a foundation for ongoing technological exchanges [3] - The "SINO (China) Preprint" platform was launched to create an open scientific ecosystem, leveraging AI and innovative technologies to enhance research communication and collaboration [4] Group 3: Achievements and Collaborations - Over 130 projects were presented, including 46 domestic and 85 international projects, covering various fields such as AI, big data, communication security, green energy, and medicine, showcasing both technological advancement and commercialization potential [5] - The conference marked a significant step for China in promoting global technological collaboration and innovation transformation, contributing to a sustainable development path driven by green energy [7]