Original link: /s/ui-O4CnT_W51_zqW4krtcQ
The development of Artificial Intelligence has come to a new stage, in which AI can interact with us in deep text like a human. Among them, OpenAI's ChatGPT is a representative model. However, to realize pleasant and effective interaction with ChatGPT, we need to understand the differences between human-computer interaction and interpersonal interaction, and then take effective measures against those differences to cope with those special challenges, only then can we chat with ChatGPT pleasantly.
First, let's look at human interaction, or interpersonal interaction. Interpersonal interaction has various purposes and results, which may be information exchange, relationship building, problem solving, etc. Interpersonal interaction is a complex and in-depth process that includes verbal and non-verbal elements such as language, facial expressions, body language, emotional colors, etc. Human beings are capable of continuous learning and adaptation in the process of interaction. Humans are capable of continuous learning and adaptation during the interaction process, and can make changes to the environment, situation, and human responses. Emotions are also a very important part of the process of interpersonal interaction, and people can perceive and express a variety of complex emotions. At the same time, in interpersonal interactions, if there is a mistake in understanding or a misunderstanding, people are usually able to quickly resolve it through further communication.
Next, let's understand the interaction between humans and machines, also known as human-computer interaction. The purpose of human-computer interaction is usually more specific and goal-oriented, such as executing commands, obtaining information, and accomplishing tasks. Human-computer interaction usually involves a human communicating with a machine through some kind of interface, which may include a variety of forms such as text, images, speech, etc. For ChatGPT, only text is currently available. To date, machines are still limited in their ability to understand. They can understand and respond to specific commands or queries, but are often unable to understand vague, ambiguous, or unpredictable human input. Machines require explicit programming or powerful machine learning algorithms to learn and adapt in a limited range. In human-computer interaction, while there are advanced systems that can recognize and model simple emotional states, this capability is usually extremely limited. The ability to handle errors or ambiguities during human-computer interaction may be limited, and this needs to be taken into account at the time of design.
So, although ChatGPT has performed very well in understanding human language and generating human natural language, it is not omnipotent. There are more than 7000 kinds of human natural languages, which are not only very complex, but also rich in multiple meanings and ambiguities, which is a great challenge for ChatGPT, a text-based interactive system. For example, dealing with ambiguous or multiple-meaning sentences may cause ChatGPT a certain amount of trouble. When communicating with ChatGPT, we should use clear and unambiguous statements as much as possible and avoid words and expressions that may cause confusion.
Also, ChatGPT's contextual understanding is limited. While it can memorize previous input and generate responses accordingly, ChatGPT may have difficulty if a conversation is too long or complex. To solve this problem, we try to keep conversations coherent, concise, and focused on the point, and don't expect ChatGPT to fully understand a long conversation or remember all the details.
Also, we must recognize that the content generated by ChatGPT may sometimes deviate from our expectations. Since ChatGPT is a statistically based method of generating text, it may sometimes produce inaccurate or irrelevant content. When this happens, we need to look at it with a critical mind and not simply or blindly accept the machine-generated content.
In addition, we need to realize that interaction with ChatGPT is entirely text-based; it cannot handle graphics or visual information. This means that if the problem we need to deal with involves visual information, ChatGPT may not be able to provide much help.
Finally, due to the limitations of the system implementation, there is a limit to the total number of Token that ChatGPT can input and output. This means that for some questions that require a long length, ChatGPT may not be able to give a complete answer. In this case, we can try to break the question into smaller parts and ask them in steps.
According to our experience in software development using ChatGPT, a structured approach to describing the problem to ChatGPT can fully utilize ChatGPT to get a more accurate, efficient, suitable and valuable solution, as shown in the figure below:
To describe the problem in a structured way, we should follow the following seven steps:
(i) Identify the core (nucleus) of the problem:
Begin by identifying the key points of the problem, including the specific issues you want to address and the goals you expect to achieve.
② Break down the problem (in detail):
Breaking down the problem into smaller, more manageable parts. This helps to get a clearer picture of the various aspects of the problem and how they relate to each other.
(iii) Provide background information (context):
Giving background information and context related to the problem helps ChatGPT better understand the actual context and needs of the problem.
④Set the priority (Priority):
Prioritize the parts of the question so that ChatGPT can provide targeted answers based on your needs and concerns.
⑤ Ask specific questions (concrete):
In describing the problem, try to use clear and specific language. Avoid the use of vague or polysemous terms to minimize the possibility of ambiguity and misunderstanding.
(vi) Stating assumptions or constraints (limitations):
If the problem involves specific assumptions or constraints, please express them clearly. This will help ChatGPT to provide solutions that are closer to the actual needs.
⑦ Specify the desired output format (output):
Clearly state the form of answer you expect, such as a list, paragraph, chart, etc. This can help ChatGPT better meet your expectations.
Overall, while there are some challenges in interacting with ChatGPT, as long as we are aware of them and can adopt appropriate strategies to deal with them, then we can effectively utilize and even navigate this powerful tool. We hope that these experiences will help make chatting with ChatGPT more enjoyable for you.
Orange is giving away books!
Like and comment on this blog, and five winners will be randomly selected in the comments section to have their technical books shipped to their home for free.To learn more about the story of payments, read "Payments in a Book
Understanding Payments in One Book makes you one of the first people to understand payments completely! Iconic book in the field of payment, payment leaders in China, the United States, Japan and other 4 countries 30 years of experience, China UnionPay Executive Vice President recommended, 360 ° interpretation of the payment. Discounted book purchase link (6.5% off):Click to learn
