What is the difference between collecting and collecting? Although there is only one word difference between these two words, collecting and collecting, there are obvious differences in usage scenarios and understanding directions. After in-depth analysis, it is not difficult to find the subtle differences. collect
Wang Ning: We’ve come to Tongtong’s town again. “Tongtong” is six years old.
"Tongtong" created by Zhu Songchun's team is not only a digital image, but also the world's first general artificial intelligence agent with value-driven and embodied growth as its core. Its mental level is equivalent to that of a five- or six-year-old child, and every time it grows, it explores the path for the transformation of the AI research paradigm.
Zhu Songchun, 57 years old, is a dual-employed professor at Peking University and Tsinghua University and the director of the Beijing Institute of General Artificial Intelligence. He has been conducting academic research in the field of artificial intelligence for more than thirty years. In the past five years, the birth of "Tongtong" began with his core exploration: establishing a heart for machines. He believes that in order to achieve a qualitative change in artificial intelligence from "completing specific tasks" to "understanding the human world", it is necessary to get rid of dependence on massive data and instead build a cognitive architecture with causal reasoning and self-awareness.
Wang Ning: Most people think that as long as you give it enough data to know all life scenarios or give it enough information, it seems that it can learn.
Zhu Songchun: Weak physical interaction and weak social interaction can be solved through big data, but for strong interaction, such as making glutinous rice balls, what does the sensor of the robot look like? How should the hand movements be done to collect this data? This is not easy to implement by default data as before. Because the experience of our communication process is different every time, when it enters thousands of households and industries, the magnitude of this scene will be too great.
Zhu Songchun: As a general intelligent agent, we actually do not do a lot of data training, so we have taken a technical route driven by cognition and value causality.
In 2010, Zhu Songchun was a professor at the University of California, Los Angeles. In 2011, during the Fukushima nuclear leakage incident, Japanese robots were unable to complete core tasks due to lack of autonomous intelligence, which inspired him to start thinking about the limitations of traditional AI research paths.
After seeing the limitations of the traditional artificial intelligence route, Zhu Songchun began to turn to general artificial intelligence research that is closer to human cognition. In 2020, he ended his overseas life and returned to China to take the lead in preparing for the establishment of the Beijing Institute of General Artificial Intelligence.
Zhu Songchun: We were thinking about whether we could build an AI baby, and later we called it “Tongtong” and regarded it as a representative of one of the roads leading to AGI.
Wang Ning: What exactly is the concept of general artificial intelligence? There is a difference between artificial intelligence and general artificial intelligence.
Zhu Songchun: There are differences. First, it has to complete unlimited tasks. Second, it has autonomy and self, and it is not something you say or do. Third, it must be driven by one's own inner value, and it must have original intention and value.
In order to let "Tongtong" truly grow up, Zhu Songchun and his team built a realistic 3D virtual world for it, allowing it to explore the world like a human child in a highly simulated physical and social environment, learn basic life skills and social skills, and develop cognition and mind at the same time.
Zhu Songchun: We refer to developmental psychology and children’s growth scales. From the age of 1 to 3, children basically learn embodied physical intelligence, which is to rummage through boxes, run around, and knock things over.
Zhu Songchun: After the age of 3 or 4, you start to learn social intelligence. It knows that you are unhappy and hate it. It needs to control itself and cannot yell or run around casually. This is when social intelligence begins to appear, and it knows how you see it. This training is social intelligence, which means that based on its cognitive architecture, it should map your feelings into its head, so that it can think from your perspective.
Wang Ning: But as a robot, you have to cultivate it and train it to produce such instincts.
Zhu Songchun: On the surface, this is called training, but in reality, there must be a structure in its head to train it. Therefore, the first step in social intelligence is that it needs to know what you are thinking, and it will begin to recognize the value of others. Another step is our common interests. When we open a company, how will the interests be distributed? We have common interests, shares, and the construction of a common value system for the country, nation, and mankind. On this basis, we assign various levels of value to it.
Zhu Songchun: Only after assigning value, everything will have its own values. Therefore, the process of training "Tongtong" is the same as the moral education we teach children, which is to build children's values and constantly upgrade them.
Now "Tongtong"'s mind has grown to the level of a six-year-old child. Its activity space has gradually expanded from a small house to a complete virtual town. In the virtual town, Tongtong will encounter some unexpected situations. At this moment, someone asks it to help find a cat feeder, and at the same time someone asks it to do other things.
Technical staff: In fact, "Tongtong" has the task management ability. When we have different tasks, it will sort according to the importance and priority of the problem and its own value to choose what it should do.
When "Tongtong" learns to say "no", it means that it is no longer just a program that passively executes instructions, but an intelligent agent with independent will and value judgment. Zhu Songchun believes that the hallmark of general artificial intelligence is the ability to complete unlimited tasks, the ability to make independent decisions, and is driven by inner values rather than passively executing instructions. "Tongtong" shows the "childhood" of general artificial intelligence.
Zhu Songchun: When it comes to the safety of artificial intelligence, we hope that its value system will be safe during the shaping process of "Tongtong". So what kind of value system is safe? A very important thing - it must be kind and have a basic conscience. At the same time, it must have a so-called cognitive immune system and a value immune system. It does not accept some values or these things. In essence, I realize that it is ultimately a cultural competition.
In the global artificial intelligence competition, hardware such as computing power and electricity are important competitive factors. However, Zhu Songchun believes that it is not only these external conditions that determine the future, but also the battle to build the underlying intelligent architecture from the source with conscience and value immunity as the core. Such an intelligent architecture can prevent AI from becoming a "bottomless tool" and build an alignment of AI safety and ethics from the underlying architecture.
Zhu Songchun’s relationship with artificial intelligence began in the 1980s. At that time, he was studying as an undergraduate in the Department of Computer Science and Technology at the University of Science and Technology of China. At a time when many people did not know what AI was, he established general intelligence as the direction of his exploration.
Zhu Songchun: My entry point is mainly academic curiosity and appeal. I think this is so interesting. I have to define and think about so many problems and realize them. I can realize the world's first general artificial intelligence. Life is so short, and it is so great that this can be done.
Having worked in this field for many years, Zhu Songchun realized that general artificial intelligence has left the laboratory and is the atomic bomb of the future information space and the winner in competition with major powers. As a member of the National Committee of the Chinese People's Political Consultative Conference, he focuses on original architecture, AI security governance, and talent cultivation. He also hopes that the growing "Tongtong" will be able to enter thousands of households and integrate into thousands of industries in the future.
Zhu Songchun: "Tongtong" is actually a brain. Hubei and central and western Chongqing need it very much. We combine the advantages of Beijing's scientific and technological innovation with their advantages of industrial innovation, and promote the application of these in the real economy, which I think is another huge advantage of China, such as emergency management and medical care. We have an intelligent agent factory dedicated to building this kind of platform. During the "15th Five-Year Plan" period, we will further form a large platform to help them build such intelligent agents.
Wang Ning: In the future, we can see "Tongtong" who are emergency managers, "Tongtong" who are doctors, and "Tongtong" who are housekeepers?
Zhu Songchun: Yes, thousands of industries and households need to do this step by step.
Producer丨Liu Bin Shen Gongfu
During the National Two Sessions in 2026, Hu Weiyi, a representative of the National People's Congress and secretary of the Party branch of Lailong Village, Pailou Town, Dongbao District, Jingmen City, Hubei Province, suggested promoting the AS700 manned airship industry to be piloted in the country and accelerating the expansion of application scenarios.
AS700 manned airship. Picture/video screenshots
In recent years, Jingmen City, Hubei Province has made every effort to promote the research and development and industrialization of the AS700, China's first large-scale civilian manned airship with completely independent intellectual property rights. At present, the project has successfully passed the acceptance of relevant scientific research projects of the Ministry of Industry and Information Technology, and its technical maturity and reliability have been verified at the national level.
"AS700 is a major breakthrough made by my country's aviation industry in the field of aerostats. Its successful development marks that my country has reached the world's advanced level in this field." Hu Weiyi said that AS700 has significant advantages such as high safety, economical operation, flexible take-off and landing, and green environmental protection. It has broad application prospects in diversified scenarios such as low-altitude tourism, scenic spot sightseeing, emergency rescue, and aerial inspection.
He said that since the launch of AS700, its innovative design concept and wide application potential have continued to attract public attention, forming a strong brand effect and market expectations, laying a good cognitive foundation for industrialization. At the same time, due to its groundbreaking significance, the product has been intensively reported by CCTV, People's Daily and many mainstream media, showing its great potential and hot prospects as a new carrier of low-altitude economy.
However, although AS700 has received national project support and enjoys high market popularity, and orders have covered many pioneering provinces in China, its industrialization success ultimately depends on sustainable commercial applications. The most prominent contradiction currently lies in the serious mismatch between the product’s popularity and technological maturity and the actual application and promotion speed in the province.
In order to solve the problem of application scenario implementation, firmly root the national technical achievements and market popularity of AS700 in Hubei, and transform it into the first-mover advantage and practical productivity of Hubei Province's low-altitude economic development, Hu Weiyi put forward the following three suggestions.
First, it is recommended that the AS700 be included in the national low-altitude economic development and key cultural tourism project planning, and that priority be given to laying out commercial operation demonstration routes in the "Two Rivers and Two Mountains" (Yangtze River, Hanjiang, Wudang Mountains, Shennongjia) golden tourist belts and key ecological functional areas, so as to create a nationally renowned "Hubei model" for low-altitude applications.
Second, it is recommended to encourage state-owned enterprises and platform companies to take the lead in demonstrating and actively introduce and operate AS700 through procurement, leasing or cooperative operations, quickly activate the initial market, verify the business model, and play a leading role.
Third, increase publicity, promotion and market cultivation, use national platform resources to focus on promoting demonstration projects, hold high-profile industry forums or application observation activities, attract the attention of industries across the country, effectively transform "media enthusiasm" into "investment enthusiasm" and "industry enthusiasm", and promote Hubei Province to become the leader and main beneficiary of the new low-altitude economic track.
Jiupai News special correspondent Zeng Yiting and Wang Jiaqing reported from Beijing
Editor Xiao Jiewu Feifei
Although there is only one word difference between these two words, collect and collect, there are obvious differences in usage scenarios and understanding directions. After in-depth analysis, it is not difficult to find the subtle differences.
Collecting mainly emphasizes the action of "collecting". It focuses on the process of gathering and storing existing items, information, etc. Whether it is physical objects or abstract concepts and data, once they form a whole and are included in a specific scope or system, it can be called a collection. This process is relatively passive and more often involves organizing and integrating existing resources.
Collecting, on the other hand, focuses on the action of "searching". It represents the process of actively seeking and exploring the unknown, with the purpose of discovering and acquiring resources that have not yet been collected or are outside the known range. Collected objects often need to be found through investigation, exploration, excavation, etc. This process is more active and proactive, emphasizing the initiative and exploratory nature of search and discovery.
Going one step further, we introduced the concept of "collection". "Collection" has the characteristics of collecting and gathering at the same time. It covers the dual meanings of active search and passive storage. Collection is usually used in situations such as the acquisition of natural resources, the collection of data, or the collection of cultural elements, reflecting human exploration and utilization of the environment, information, or culture.
To sum up, although the words "collect" and "collect" are exactly the same, the logic and application scenarios behind them are completely different. Collecting focuses on the organization and integration of existing resources, while collecting emphasizes the process of actively exploring and acquiring unknown resources. At the same time, the concept of "collection" further enriches our understanding of resource acquisition methods and reflects the diversity and flexibility of human beings in resource exploration and utilization in different fields.
On the evening of March 6, Liu Qingfeng, deputy to the National People's Congress and chairman of iFlytek, was interviewed by the media at the Anhui News Center of the 2026 National Two Sessions in Beijing.
Talking about the way for China's large-scale model companies to break through against the background of increasingly fierce global artificial intelligence competition and facing multiple challenges from technical limitations and industrial dimensions, Liu Qingfeng said that the key lies in achieving full-stack independent controllability from algorithms, data to application scenarios.
Liu Qingfeng, deputy to the National People's Congress and chairman of iFlytek. Picture/Jiupai News
He pointed out that many models are currently trained based on foreign computing power and do adaptation and inference research on domestic computing power. This model still has security risks at the autonomous and controllable level. "A truly autonomous and controllable model should be trained on one's own computing power. Only after training on one's own computing power can a more complete autonomous ecosystem be formed. I think our safe future is guaranteed."
In his view, domestic computing power training has made continuous progress in the past year. "Although Spark is still the only mainstream model for domestic computing power training today, more and more special models and model manufacturers have begun to adapt artificial intelligence to domestic computing power, and have also begun to enter the training of base models."
Based on these developments, he showed firm confidence: "China's own computing power continues to develop, algorithms continue to adapt, and we are making our own innovations. Coupled with our rich application scenarios and the advantages of software and hardware integration, China's artificial intelligence can not only achieve independent controllability during the '15th Five-Year Plan', but also has a very broad future."
Jiupai News special correspondent Li Kai and Yan Huayang reported from Beijing
Editor Xiao Jie Ren Zhuo
China Jiangxi Net News (Reporter Chen Zhang) The modern logistics industry is an important pillar of our province’s modern service industry and one of the outstanding shortcomings in economic and social development. Recently, our province formulated the "Jiangxi Province Promotes the Implementation of "Internet +" Efficient Logistics Work Plan", striving to achieve more than 50% of the application rate of advanced information technology in the logistics field by 2020, and the added value of the logistics industry generated by new logistics technologies, new models, and new formats based on the Internet will account for about 40% of the added value of the entire industry.
The "Plan" points out that it is necessary to improve and improve the provincial logistics public information platform, promote the construction of provincial comprehensive transportation information interconnection and exchange nodes, and establish railway, waterway, highway, aviation, postal and other logistics information exchange nodes in the province's key first- and second-level railway logistics bases, Nanchang Port Longtougang Comprehensive Hub, Jiujiang Port Hongguang Hub and other important water transport hubs, Nanchang Changbei International Airport, Ganzhou Golden Airport and other major airports, and large comprehensive logistics parks.
In terms of smart logistics, promote new models of instant delivery. Promote the coordinated development of e-commerce and logistics, promote community terminal distribution models such as "order online and pick up in stores" and "smart pickup cabinets", and accelerate the construction of e-commerce logistics systems such as express logistics parks, parcel sorting centers, and terminal distribution outlets to achieve same-day orders and next-day delivery. At the same time, we will study and improve the "car-free carrier" management system, and encourage and guide well-known enterprises in the industry such as Provincial State-controlled Logistics Company, Wanji Logistics Company, and Zhengguangtong Supply Chain Management Company to carry out "car-free carrier" pilot work.
Regarding the creation of logistics outlets, our province will plan to layout 1 first-level logistics base, 6 second-level logistics bases, and 10 third-level logistics bases. Based on the third-level railway logistics base network, we will build logistics operation stations, acceptance stations, acceptance points and trackless stations as needed to create a distinct layered railway logistics park network. In addition, our province will also support the provincial state-controlled logistics company to initiate the establishment of the Jiangxi "Internet +" Logistics Enterprise Alliance by means of a modern information platform, guide and encourage logistics enterprises in the province to join the logistics alliance, and form a Jiangxi logistics backbone network as soon as possible.
1. The meanings of “collect” and “collect” are respectively:
1. "Collect": Search, collect (Gather), collect data from all processes in the group to one process/find everywhere (things are gathered together).
2. "Collect": to recruit and gather; to collect, collect and bring together scattered things.
2. Difference:
1. Compared with collecting, collecting has certain selectivity and directionality.
2. Collect means to look for things and then gather them together. Collect means to gather things directly without searching.
For example, "collecting stamps" may mean that there is a certain stamp on the market, and you just go and buy it; while "collecting stamps" means that a certain rare stamp is not available on the market, and you need to search and buy it. You can even think of collection = search + collection.
3. Different pinyin:
collect (sōu jí)
collect (shōu jí)
4. Antonyms:
Antonyms for collection: distribution;
Antonyms for collection: scatter, distribute;
5. Basic meaning:
Collect: search, collect;
Gather: to gather together;
6. English translation:
Collect: collect gather;
Collect: to collect, to get together, to gather, to glean;
Extended information:
collect:
1. Recruitment gathering.
"Book of the Later Han·Biography of Zhang Huan": "Huan didn't listen, so he marched to the Great Wall and gathered soldiers." (Zhang Huan didn't listen, so he marched to the Great Wall and gathered soldiers.)
2. Collect, collect, and put together scattered things.
Qin Mu's "Flower City: Collecting Shells on the Beach": "There are some 'shell fans' who specialize in collecting shells."
collect:
1. Search collection.
Qing Dynasty Shen Fu's "Six Notes of a Fusheng·Boujingjile": "Books that are incomplete and incomplete must be collected into separate categories and compiled into volumes, collectively called 'broken slips and residual editions'." (Books that are incomplete and incomplete must be collected into separate categories and compiled into volumes, collectively known as 'broken slips and residual editions'.)
Tian Beihu's "On the Origin of Articles": "Collecting ballads and proverbs will reveal the origin of "Erya"; including summer and winter, it is the secret key of the word." (Collecting ballads and proverbs will reveal the origin of "Erya"; including summer and winter, then the secret key of these words.)
Ba Jin's "Autumn" 2: "We have already divided the calligraphy and paintings that Dad worked so hard to collect."
2. Collect: Collect.
"Preface" by Zeng Guofan of the Qing Dynasty: "There are very few articles that are based on textual research. The advocacy of previous philosophers is not ambitious, and the admiration of later generations is also small." (The articles that are only based on evidence are very few, the advocacy of previous philosophers is not ambitious, and the admiration of later generations is also small.)
Reference: Baidu Encyclopedia-Collect
Baidu Encyclopedia-Collection
A Complete Analysis Of The Basic Concepts Of Stock Restoration And The Calculation And Application Of Pre-restoration2026-03-03Comments(0)
21st Century Economic Report2026-02-23Comments(0)
In-depth Analysis Of The Core Characteristics, Price Performance, And Future Trends And Opportunities Of Pepe Currency2026-02-24Comments(0)