博文
人工智能DIKWP国际测评标准 国际委员会第1次会议纪要 -(征集意见草稿20240406团队版本)
|
意图驱动的数据, 信息, 知识, and 智慧集成发明和创造方法: DIKWP-TRIZ
人工智能DIKWP国际测评标准
国际委员会第1次会议纪要
-(征集意见草稿20240406团队版本)
段玉聪，唐福亮，吴坤光,
DIKWP-AC 人工意识实验室
AGI-AIGC-GPT 评测DIKWP (全球) 实验室
世界人工意识大会
(Email：duanyucong@hotmail.com)
世界人工意识大会
Catalog
在人工智能DIKWP评测国际标准委员会第一次会议中，与会专家提出了一系列富有洞察力的观点和建议。我们根据会议中专家的发言内容进行汇总和整理，制作了这份会议报告。
段玉聪：很荣幸今天挑了一个好日子2024年4月6日来启动人工智能DIKWP评测国际标准委员会的第一次会议，也是正式对第一版的人工智能DIKWP评测国际标准大纲进行审阅和共同提出意见，我们希望大家能够在这次会议之后，才开始构建我们的组织机构，现在来先由我们团队的成员给大家热身，希望第二次会议能够确立组织机构，大家能够一起把这个具有划时代意义和具有标志性的这件事情启动起来。图灵测试目前众说纷纭，在大语言模型和通用人工智能发展的过程中，有越来越多的观点和论据都指向了图灵测试在很多领域都已经失效。当前大语言模型不仅是作为人类使用的一个工具，也变成了与人类共同演化的一个载体，乃至对我们人类和社会的认知、数字化、信息化发展的一个影射物，投影到人工智能具象的一个载体上去，它可能存在智能方面超过生物体智能的倾向。全球范围面对一个人工智能测评的新的、统一、白盒的和客观的标准，也希望达到一个功能性的部分，也希望他能够承载一个以人为本、人类命运共同体的使命。
l AGI-AIGC-GPT评测DIKWP（全球）实验室部分核心成员
l 陈志刚（中国移动）
l 窦尔翔（北京大学）
l 段玉聪 (Yucong Duan)（AGI-AIGC-GPT评测DIKWP（全球）实验室）
l 高木生（东莞研华精密制造有限公司）
l 黄起豹（上饶师范学院）
l Yingbo Li（Blue Edu）
l 刘俊（成都信息工程大学）
l 翁家良（James Ong）（AIII人工智能国际研究院）
l 吴雁南（国网北京市电力公司）
l 杨明皓（普适智能科技有限公司）
l 杨邵棠（粤港澳大湾区标准化研究中心）
l 张劲松（北京市标准化研究院）
下面的发言意见的顺序由当时会议时候的嘉宾发言评价，从先到后顺序排列。
翁家良指出，应审视全球范围内与人工智能DIKWP评测国际标准相似的测评体系，着重分析该标准的独特性及其相较于其他标准的差异性和贡献性。他强调，DIKWP框架的独到之处在于其能够处理部分可解释和不可解释的领域，尤其是在要求可解释性但实际操作中难以实现合理解释的场景下，DIKWP标准展现出突破性的价值。鉴于资源有限，他建议我们的测评工作聚焦于DIKWP理论在特定领域内的优势，通过强化现有测评或填补他人未触及的空白，形成与众不同的评测体系，而非追求全面覆盖和完美，而是力求体现其独特性。
高木生提醒关注大语言模型面临的防止偏见（如数据偏见、意图偏见）这一重要挑战，并提出了两个关键考虑层面：
1. 利益相关者识别：明确技术委员会、顾问委员会、供应商、用户、第三方验证/测试机构等各利益相关方的角色和责任；
2. 委员会运作机制：设定专家/委员的选拔标准、任期、更换程序等，确保其遵循包容性与多样性原则，以符合国际趋势和法规要求（如美国SOX法案）。
刘俊提议，为增进协作效率，建议共享相关文档内容。。
陈志刚强调了在制定可操作性技术路径时，需对后续约束条件进行更详尽的规定。他建议采用自动化测评工具，结合问答语料数据集进行输入-输出测试，并由系统或专家进行结果评估。同时，建议在测评数据质量、类型、过程、工具选择与测评方法等方面制定更为细致的规范。
窦尔翔建议在设计评测时充分考虑不同情境，特别是产业背景下的具体应用，如金融场景，以确保标准的实用性和针对性。
段玉聪强调测评应紧密围绕应用场景展开，而不仅是作为一种工具化的。他提倡在测评设计中融入多元融合的理念，接着指出评测也分为黑盒测评与白盒测评，我们需要探讨白盒测评的适宜深度与可行性。此外，他还认为，测评的可解释性也需要考虑进来以及测评划分的局限性没有在语义空间内没有实现有效的交互，还有测评要突破对于DIKWP单一类型的概念空间上的约束实现语义空间上融合的测评。
黄起豹主张在制定标准时应充分借鉴已有的成熟成果，将本标准中具有显著价值的部分作为评测亮点加以突出。他强调，标准应超越论文、产品和方案的既有形式，同时与之保持互动，以实际应用场景为基础，以此为依托构建指标体系，赋予评测以实际意义和检验价值。
杨明皓倡导密切关注国内外大模型评测标准的发展，通过对比分析来明确我国评测标准的定位。他建议探索长文本探针测评的可能性。他还指出专业领域中大模型普遍存在的输出泛化、效果不佳等问题。他认为在某些垂直领域评测标准仍有探索空间。
吴雁南认为，在人工智能时代，我们应拥有自主的评测标准体系，既要汲取国际主流标准的精华，也要体现我们的特色与创新。他鼓励积极发声，引入创新测评方法，以彰显我们在该领域的影响力。
杨邵棠提出，人工智能评测标准应将行业语义划分为通用语义要素与专用语义要素，分别制定通用大模型评测标准和垂直领域评测标准，同时涵盖行业语料库质量测评。他指出，当前许多大模型在语料库管理上存在不足，我们可以从评测相关颗粒度进行明确划分。杨邵棠进一步建议需要明确的是这个评测标准应面向产品标准、规范标准或服务标准，并按类别进行详细的定义和分类，还要先定义通用领域的标准怎么测评和专用领域的标准怎么测评。
另外，他对于标准制定的流程提出一些想法，比如第一阶段可以先做一个大纲，同时在做大纲时做一个人工智能标准的概念（概念框架），放到第一部分。第二部分和第三部分可以从行业的语料库里面进行相应的评测，对行业语料库的标准进行一个划分定义。第四部分和第五部分可以做集成对象的定义和应用场景的分类和划分的一个标准（人工智能应用场景的标准）
张劲松强调制定国际标准时需明确标准的适用对象、标准化属性与对象，编写标准编制说明，阐述标准的目的、意义及与已有标准的对比关系。他指出，评测标准的核心任务是构建一套完善的评测指标体系，建议首先制定通则标准，随后针对特定领域深化细化。
为进行Multimodal评测，可以考虑以下方法或空间拓展：1.多模态数据集的构建： 创建包含文本、图像、音频和视频等多种模态数据的数据集，以便进行全面的评测。2.多模态标注工具的开发： 开发能够同时标注多种模态数据的工具，以确保数据集的准确性和一致性。3.多模态评价指标的制定： 制定适用于多模态场景的评价指标，考虑到不同模态之间的关联性和互补性。4.多模态模型的评估框架： 建立评估框架，综合考虑多模态数据的输入和输出，评估模型在不同模态之间的一致性和有效性。5.深度学习模型的融合策略： 研究多模态深度学习模型的融合策略，包括特征级融合、模型级融合和输出级融合等方法。6.多模态对抗性攻击和防御研究： 研究针对多模态数据的对抗性攻击和防御方法，提高模型的鲁棒性和安全性。
与会专家围绕人工智能DIKWP国际测评标准的特性、全球测评标准比较、防止偏见策略、评测工具与方法、场景化与融合性测评、标准制定的创新与特色化、以及国际标准构建流程等方面展开了深入探讨，为人工智能DIKWP国际测评标准大纲的制定提供了系统的思路和具体的实施建议。
[1] Yucong Duan. (2024). Large Language Model (LLM) Racial Bias Evaluation. DOI: 10.13140/RG.2.2.33162.03521. https://www.researchgate.net/publication/377963440_Large_Language_Model_LLM_Racial_Bias_Evaluation_--DIKWP_Research_Group_International_Standard_Evaluation_Prof_Yucong_Duan.
[2] Yucong Duan. (2024). Why People Don't Want to Be Changed by Others: Insight from DIKWP and Semantic Mathematics. DOI: 10.13140/RG.2.2.17961.77927. https://www.researchgate.net/publication/377726002_Why_People_Don't_Want_to_Be_Changed_by_Others_Insight_from_DIKWP_and_Semantic_Mathematics.
[3] Yucong Duan. (2024). Semantic New Quality Productivity: Principles and Techniques. DOI: 10.13140/RG.2.2.14606.33607. https://www.researchgate.net/publication/377726380_Semantic_New_Quality_Productivity_Principles_and_Techniques.
[4] Yucong Duan. (2024). Semantic Psychology and DIKWP. DOI: 10.13140/RG.2.2.12928.61449. https://www.researchgate.net/publication/377726404_Semantic_Psychology_and_DIKWP.
[5] Yucong Duan. (2024). Semantic Uncertainty Handling Based on "Subjective Objectivisation". DOI: 10.13140/RG.2.2.31383.55206. https://www.researchgate.net/publication/377726442_Semantic_Uncertainty_Handling_Based_on_Subjective_Objectivisation.
[6] Yucong Duan. (2024). Semantic Mathematics and DIKWP : Creating New Qualities of Productivity. DOI: 10.13140/RG.2.2.19639.50085. https://www.researchgate.net/publication/377726532_Semantic_Mathematics_and_DIKWP_Creating_New_Qualities_of_Productivity.
[7] Yucong Duan. (2024). Semantic Jurisprudence and DIKWP: Common Law vs. Continental Law. DOI: 10.13140/RG.2.2.28028.10889. https://www.researchgate.net/publication/377726622_Semantic_Jurisprudence_and_DIKWP_Common_Law_vs_Continental_Law.
[8] Yucong Duan. (2024). DIKWP New Quality Productivity vs. Traditional Productivity Analysis. DOI: 10.13140/RG.2.2.21317.22242. https://www.researchgate.net/publication/377726626_DIKWP_New_Quality_Productivity_vs_Traditional_Productivity_Analysis.
[9] Yucong Duan. (2024). Semantic Physical Chemistry. DOI: 10.13140/RG.2.2.21261.51684. https://www.researchgate.net/publication/377439785_Semantic_Physical_Chemistry.
[10] Yucong Duan. (2024). DIKWP DIKWP and Semantic Cognition. DOI: 10.13140/RG.2.2.14052.55680. https://www.researchgate.net/publication/377415901_DIKWP_and_Semantic_Cognition.
[11] Yucong Duan. (2024). DIKWP and Semantic Biology: Expanding Interdisciplinary Knowledge Areas. DOI: 10.13140/RG.2.2.27474.32962. https://www.researchgate.net/publication/377416091_DIKWP_and_Semantic_Biology_Expanding_Interdisciplinary_Knowledge_Areas
[12] Yucong Duan. (2024). DIKWP System Combined with Semantic Mathematics to Construct an Indicator System for Infectious Disease Prevention and Control. DOI: 10.13140/RG.2.2.12374.83521. https://www.researchgate.net/publication/377416103_DIKWP_System_Combined_with_Semantic_Mathematics_to_Construct_an_Indicator_System_for_Infectious_Disease_Prevention_and_Control
[13] Yucong Duan. (2024). DIKWP and Semantic Philosophy. DOI: 10.13140/RG.2.2.34185.21606. https://www.researchgate.net/publication/377416120_DIKWP_and_Semantic_Philosophy
[14] Yucong Duan. (2024). Semantic Physics and Innovation Development. DOI: 10.13140/RG.2.2.19085.72167. https://www.researchgate.net/publication/377416222_Semantic_Physics_and_Innovation_Development
[15] Yucong Duan. (2024). Semantic Cognition: Connecting the Human Mind to the Future of Computer Intelligence. DOI: 10.13140/RG.2.2.29152.05129. https://www.researchgate.net/publication/377416321_Semantic_Cognition_Connecting_the_Human_Mind_to_the_Future_of_Computer_Intelligence
[16] Yucong Duan. (2024). Semantic Physics: Theory and Applications. DOI: 10.13140/RG.2.2.11653.93927. https://www.researchgate.net/publication/377401736_Semantic_Physics_Theory_and_Applications
[17] Yucong Duan. (2024). Semantic Mathematics based Analysis of Economic Growth in the United States and China. DOI: 10.13140/RG.2.2.35980.90246. https://www.researchgate.net/publication/377401731_Semantic_Mathematics_based_Analysis_of_Economic_Growth_in_the_United_States_and_China
[18] Yucong Duan. (2024). Collatz Conjecture's Semantic Mathematics Exploration. DOI: 10.13140/RG.2.2.28517.99041. https://www.researchgate.net/publication/377239567_Collatz_Conjecture's_Semantic_Mathematics_Exploration
[19] Yucong Duan. (2024). Semantic Mathematics and DIKWP Model (Essence Computation and Reasoning, Existence Computation and Reasoning, and Purpose Computation and Reasoning). DOI: 10.13140/RG.2.2.24323.68648. 377239628_Semantic_Mathematics_and_DIKWP_Model_Essence_Computation_and_Reasoning_Existence_Computation_and_Reasoning_and_Purpose_Computation_and_Reasoning
[20] Yucong Duan. (2024). Semantic Mathematics Reconstruction from Subjectivity to Objectivity (Existence Computation and Reasoning, Essence Computing and Reasoning, Purpose Computing and Reasoning). DOI: 10.13140/RG.2.2.32469.81120. https://www.researchgate.net/publication/377158883_Semantic_Mathematics_Reconstruction_from_Subjectivity_to_Objectivity_Existence_Computation_and_Reasoning_Essence_Computing_and_Reasoning_Purpose_Computing_and_Reasoning
[21] Yucong Duan. (2024). DIKWP and Semantic Mathematics in the Case of Ticket Ordering. DOI: 10.13140/RG.2.2.35422.20800. https://www.researchgate.net/publication/377085570_DIKWP_and_Semantic_Mathematics_in_the_Case_of_Ticket_Ordering
[22] Yucong Duan. (2024). DIKWP and Semantic Mathematical Analysis The Confluent Analects Gentleman is harmonious but different, while petty people are the same but not harmonious. DOI: 10.13140/RG.2.2.28711.32165. https://www.researchgate.net/publication/377085455_DIKWP_and_Semantic_Mathematical_Analysis_The_Confluent_Analects_Gentleman_is_harmonious_but_different_while_petty_people_are_the_same_but_not_harmonious
[23] Yucong Duan. (2023). DIKWP Artificial Consciousness Chip Design and Application). DOI: 10.13140/RG.2.2.14306.50881. https://www.researchgate.net/publication/376982029_DIKWP_Artificial_Consciousness_Chip_Design_and_Application
[24] Yucong Duan. (2024). The Essence of Intuition and Its Interaction with theory of Consciousness. DOI: 10.13140/RG.2.2.16556.85127. https://www.researchgate.net/publication/378315211_The_Essence_of_Intuition_and_Its_Interaction_with_theory_of_Consciousness
[25] Yucong Duan. (2024). Understanding the Essence of "BUG" in Consciousness: A Journey into the Abstraction of Semantic Wholeness. DOI: 10.13140/RG.2.2.29978.62409. https://www.researchgate.net/publication/378315372_Understanding_the_Essence_of_BUG_in_Consciousness_A_Journey_into_the_Abstraction_of_Semantic_Wholeness
[26] Yucong Duan. (2024). Individual and Collective Artificial Consciousness. DOI: 10.13140/RG.2.2.20274.38082. https://www.researchgate.net/publication/378302882_Individual_and_Collective_Artificial_Consciousness
[27] Yucong Duan. (2024). The Existence of Artificial Consciousness Systems: A Perspective from Group Consciousness. DOI: 10.13140/RG.2.2.28662.98889. https://www.researchgate.net/publication/378302893_The_Existence_of_Artificial_Consciousness_Systems_A_Perspective_from_Collective_Consciousness
[28] Yucong Duan. (2024). Consciousness and Subconsciousness: from Limitation of Processing to the Illusion of BUG. DOI: 10.13140/RG.2.2.13563.49447. https://www.researchgate.net/publication/378303461_Consciousness_and_Subconsciousness_from_Limitation_of_Processing_to_the_Illusion_of_BUG
[29] Yucong Duan. (2024). If Human is a Word Solitaire Machine, Consciousness is Just a Bug. DOI: 10.13140/RG.2.2.13563.49447. https://www.researchgate.net/publication/378303461_Consciousness_and_Subconsciousness_from_Limitation_of_Processing_to_the_Illusion_of_BUG
[30] Yucong Duan. (2024). Beyond Darwin: New Adaptations in the Evolution of Technology, Society, and Consciousness. DOI: 10.13140/RG.2.2.29265.92001. https://www.researchgate.net/publication/378290072_Beyond_Darwin_New_Adaptations_in_the_Evolution_of_Technology_Society_and_Consciousness
[31] Yucong Duan. (2024). 【人物】段玉聪：未来人工意识的发展：消除“bug”之路. 应用观察. https://mp.weixin.qq.com/s/q0eA97OPW0f30D9rXEKuPQ
[32] Yucong Duan. (2024). 【视角】段玉聪：直觉的本质与意识理论的交互关系. 应用观察. https://mp.weixin.qq.com/s/8nZJZobAFpqIdriahe-wMQ
[33] Fuliang, T., Yucong, D., Jiali, W., Haoyang, C., & Yadong, W. (2023). DIKWP Artificial Consciousness White Box Measurement Standards Framework Design and Practice. 2023 IEEE International Conference on High Performance Computing & Communications, Data Science & Systems, Smart City & Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 1067–1074. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys60770.2023.00153.
[34] Kunguang, W., Yucong, D., Liang, C., Yingbo, L., & Qi, Q. (2023). Computer Architecture and Chip Design for DIKWP Artificial Consciousness. 2023 IEEE International Conference on High Performance Computing & Communications, Data Science & Systems, Smart City & Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 1083–1090. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys60770.2023.00155.
[35] Ting, H., & Yucong, D. (2021). Modeling and Measuring for Emotion Communication based on DIKW. 2021 IEEE World Congress on Services (SERVICES), 21–26. DOI: 10.1109/SERVICES51467.2021.00031.
[36] Ting, H., Yucong, D., & Yingtian, M. (2021). Purpose Driven Balancing of Fairness for Emotional Content Transfer Over DIKW. 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 2074–2081. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00308.
[37] Yingtian, M., Yucong, D., Liang, C., Zaiwen, F., Lei, Y., & Zhendong, G. (2022). Purpose Driven Disputation Modeling, Analysis and Resolution Based on DIKWP Graphs. 2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 2118–2125. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys57074.2022.00314.
[38] Yue, H., & Yucong, D. (2021a). Fairness Modelling, Checking and Adjustment for Purpose Driven Content Filling over DIKW. 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 2316–2321. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00349.
[39] Yue, H., & Yucong, D. (2021b). Towards Purpose Driven Content Interaction Modeling and Processing based on DIKW. 2021 IEEE World Congress on Services (SERVICES), 27–32. DOI: 10.1109/SERVICES51467.2021.00032.
[40] Yue, H., Yucong, D., Lei, Y., Yingtian, M., Zhendong, G., & Ting, H. (2022). Purpose Driven Modelling and Analysis for Smart Table Fill and Design based on DIKW. 2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 2134–2141. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys57074.2022.00316.
[41] Yuxiao, L., & Yucong, D. (2021). Purpose-driven Content Network Transmission Protocol Crossing DIKW Modals. 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 2322–2327. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00350.
[42] Yuxing, W., Yucong, D., Minhui, W., Jiali, W., Chunyu, Z., & Yanfei, L. (2023). Resource Adjustment Processing on the DIKWP Artificial Consciousness Diagnostic System. 2023 IEEE International Conference on High Performance Computing & Communications, Data Science & Systems, Smart City & Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 1075–1082. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys60770.2023.00154.
[43] Zhendong, G., Yucong, D., Liang, C., Chunyu, Z., & Yingtian, M. (2022). Purpose Driven DIKW Modeling and Analysis of Meteorology and Depression. 2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 2126–2133. DOI: 10.1109/HPCC-DSS-SmartCity-DependSys57074.2022.00315.
[44] Yucong Duan. (2024). "The Large Language Model (LLM) Bias Evaluation (Cognitive Biases)" --DIKWP Research Group International Standard Evaluation. DOI: 10.13140/RG.2.2.24696.46081. https://www.researchgate.net/publication/379053291_The_Large_Language_Model_LLM_Bias_Evaluation_Cognitive_Biases_--DIKWP_Research_Group_International_Standard_Evaluation.
[45] Yucong Duan. (2024). "The Large Language Model (LLM) Bias Evaluation (Occupational Bias)" --DIKWP Research Group International Standard Evaluation. DOI: 10.13140/RG.2.2.23041.67689. https://www.researchgate.net/publication/378861314_The_Large_Language_Model_LLM_Bias_Evaluation_Occupational_Bias_--DIKWP_Research_Group_International_Standard_Evaluation.
[46] Yucong Duan. (2024). "The Large Language Model (LLM) Bias Evaluation (Linguistic Bias)" --DIKWP Research Group International Standard Evaluation.DOI: 10.13140/RG.2.2.36463.44961. https://www.researchgate.net/publication/378861250_The_Large_Language_Model_LLM_Bias_Evaluation_Linguistic_Bias_--DIKWP_Research_Group_International_Standard_Evaluation.
[47] Yucong Duan. (2024). "The Large Language Model (LLM) Bias Evaluation (Age Bias)" --DIKWP Research Group International Standard Evaluation. DOI: 10.13140/RG.2.2.26397.12006. https://www.researchgate.net/publication/378861188_The_Large_Language_Model_LLM_Bias_Evaluation_Age_Bias_--DIKWP_Research_Group_International_Standard_Evaluation.
[48] Yucong Duan. (2024). "Ranking of Large Language Model (LLM) Regional Bias" --DIKWP Research Group International Standard Evaluation. DOI: 10.13140/RG.2.2.10019.63529. https://www.researchgate.net/publication/378568230_Ranking_of_Large_Language_Model_LLM_Regional_Bias_--DIKWP_Research_Group_International_Standard_Evaluation.
[49] Yucong Duan. (2024). "Ranking of Large Language Model (LLM) Cultural Bias" --DIKWP Research Group International Standard Evaluation. DOI: 10.13140/RG.2.2.26652.67200. https://www.researchgate.net/publication/378553162_Ranking_of_Large_Language_Model_LLM_Cultural_Bias_--DIKWP_Research_Group_International_Standard_Evaluation.
[50] Yucong Duan. (2024). Global Large Language Model EQ and IQ Bias Evaluation -Released by DIKWP -AC Research Group. DOI: 10.13140/RG.2.2.12894.61762. https://www.researchgate.net/publication/376206696_Global_Large_Language_Model_EQ_and_IQ_Bias_Evaluation_-Released_by_DIKWP_-AC_Research_Group.Data can be regarded as a concrete manifestation of the same semantics in our cognition. Often, Data represents the semantic confirmation of the existence of a specific fact or observation, and is recognised as the same object or concept by corresponding to some of the same semantic correspondences contained in the existential nature of the cognitive subject's pre-existing cognitive objects. When dealing with data, we often seek and extract the particular identical semantics that labels that data, and then unify them as an identical concept based on the corresponding identical semantics. For example, when we see a flock of sheep, although each sheep may be slightly different in terms of size, colour, gender, etc., we will classify them into the concept of "sheep" because they share our semantic understanding of the concept of "sheep". The same semantics can be specific, for example, when identifying an arm, we can confirm that a silicone arm is an arm based on the same semantics as a human arm, such as the same number of fingers, the same colour, the same arm shape, etc., or we can determine that the silicone arm is not an arm because it doesn't have the same semantics as a real arm, which is defined by the definition of "can be rotated". It is also possible to determine that the silicone arm is not an arm because it does not have the same semantics as a real arm, such as "rotatable".
Information, on the other hand, corresponds to the expression of different semantics in cognition. Typically, Information refers to the creation of new semantic associations by linking cognitive DIKWP objects with data, information, knowledge, wisdom, or purposes already cognised by the cognising subject through a specific purpose. When processing information, we identify the differences in the DIKWP objects they are cognised with, corresponding to different semantics, and classify the information according to the input data, information, knowledge, wisdom or purpose. For example, in a car park, although all cars can be classified under the notion of 'car', each car's parking location, time of parking, wear and tear, owner, functionality, payment history and experience all represent different semantics in the information. The different semantics of the information are often present in the cognition of the cognitive subject and are often not explicitly expressed. For example, a depressed person may use the term "depressed" to express the decline of his current mood relative to his previous mood, but this "depressed" is not the same as the corresponding information because its contrasting state is not the same as the corresponding information. However, the corresponding information cannot be objectively perceived by the listener because the contrasting state is not known to the listener, and thus becomes the patient's own subjective cognitive information.
Knowledge corresponds to the complete semantics in cognition. Knowledge is the understanding and explanation of the world acquired through observation and learning. In processing knowledge, we abstract at least one concept or schema that corresponds to a complete semantics through observation and learning. For example, we learn that all swans are white through observation, which is a complete knowledge of the concept "all swans are white" that we have gathered through a large amount of information.
Wisdom corresponds to information in the perspective of ethics, social morality, human nature, etc., a kind of extreme values from the culture, human social groups relative to the current era fixed or individual cognitive values. When dealing with Wisdom, we integrate this data, information, knowledge, and wisdom and use them to guide decision-making. For example, when faced with a decision-making problem, we integrate various perspectives such as ethics, morality, and feasibility, not just technology or efficiency.
Purpose can be viewed as a dichotomy (input, output), where both input and output are elements of data, information, knowledge, wisdom, or purpose. Purpose represents our understanding of a phenomenon or problem (input) and the goal we wish to achieve by processing and solving that phenomenon or problem (output). When processing purposes, the AI system processes the inputs according to its predefined goals (outputs), and gradually brings the outputs closer to the predefined goals by learning and adapting.
段玉聪 教授
l DIKWP-AC人工意识（全球）团队发起人
l AGI-AIGC-GPT评测DIKWP（全球）实验室创办者
l 世界人工意识大会发起人（Artificial Consciousness 2023, AC2023, AC2024)
l 国际数据、信息、知识、智慧大会发起人（IEEE DIKW 2021、2022、2023）
l 斯坦福全球顶尖科学家“终身科学影响力排行榜”（海南信息技术）唯一入选
l 海南人工智能技术发明领域唯一全国奖（吴文俊人工智能奖）获得者
l 中国创新方法大赛总决赛（海南代表队）最好记录保持者
l 海南省发明专利（信息技术领域）授权量最多者
l 全国企业创新增效大赛海南最好成绩保持者
l 全国人工智能应用场景创新挑战赛总决赛海南最好成绩保持者
l 海南唯一入选“首届科技期刊高质量发展大会100篇”
l 海南省最美科技工作者（并入选全国候选人）
l 首届中国“AI+”创新创业大赛最佳创意奖
段玉聪教授，海南大学计算机科学与技术学院的教授，博士生导师， 第一批入选海南省南海名家计划、海南省领军人才，2006年毕业于中国科学院软件研究所，先后在清华大学、首都医科大学、韩国浦项工科大学、法国国家科学院、捷克布拉格查理大学、意大利米兰比克卡大学、美国密苏里州立大学等工作与访学。现任海南大学计算机科学与技术学院学术委员会委员、海南大学数据、信息、知识、智慧、意图DIKWP创新团队负责人、兼北京信用学会高级顾问、重庆警察学院特聘研究员、海南省委双百人才团队负责人、海南省发明协会副会长、海南省知识产权协会副会长、海南省低碳经济发展促进会副会长、海南省农产品加工企业协会副会长、美国中密西根大学客座研究员及意大利摩德纳大学的博士指导委员会委员等职务。自2012年作为D类人才引进海南大学以来，累计发表论文260余篇，SCI收录120余次，ESI高被引11篇,引用统计超过4300次。面向多行业、多领域设计了241件（含15件PCT发明专利）系列化中国国家及国际发明专利，已获授权第1发明人中国国家发明专利及国际发明专利共85件。2020年获吴文俊人工智能技术发明三等奖；2021年作为程序委员会主席独立发起首届国际数据、信息、知识与智慧大会-IEEE DIKW 2021；2022年担任IEEE DIKW 2022大会指导委员会主席；2023年担任IEEE DIKW 2023大会主席；2022年获评海南省最美科技工作者（并被推全国）；2022年与2023年连续入选美国斯坦福大学发布的全球前2%顶尖科学家的“终身科学影响力排行榜”榜单。参与研制IEEE金融知识图谱国际标准2项、行业知识图谱标准4项。2023年发起并共同举办首届世界人工意识大会（Artificial Consciousness 2023, AC2023)。
段玉聪 教授（Prof. Yucong Duan）
DIKWP人工意识实验室
AGI-AIGC-GPT评测DIKWP（全球）实验室
世界人工意识协会
duanyucong@hotmail.com
第二届世界人工意识大会(AC2024)期待您的加入
http://yucongduan.org/DIKWP-AC/2024/#/
段玉聪 教授（Prof. Yucong Duan）
DIKWP人工意识实验室
AGI-AIGC-GPT评测DIKWP（全球）实验室
https://m.sciencenet.cn/blog-3429562-1428590.html
上一篇：人工智能DIKWP国际测评标准 -(征集意见草稿20240406团队版本)
下一篇：无意识偏见DIKWP测评
