CN117474090A - Prompt word sharing method and device - Google Patents
Prompt word sharing method and device Download PDFInfo
- Publication number
- CN117474090A CN117474090A CN202311545314.5A CN202311545314A CN117474090A CN 117474090 A CN117474090 A CN 117474090A CN 202311545314 A CN202311545314 A CN 202311545314A CN 117474090 A CN117474090 A CN 117474090A
- Authority
- CN
- China
- Prior art keywords
- prompt
- word
- sharing
- sharer
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000013473 artificial intelligence Methods 0.000 claims abstract description 14
- 238000012986 modification Methods 0.000 claims description 11
- 230000004048 modification Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000007792 addition Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000012217 deletion Methods 0.000 description 3
- 230000037430 deletion Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The artificial intelligence large model inputs information which is prompt words (prompt), wherein the prompt words are composed of contents such as purposes, reference information and the like. For the multi-mode artificial intelligence large model, the generalized prompt word also comprises multi-mode information types such as pictures, voice, video and the like. Because of the nature of large models, hints of different quality will lead to a vast variance in the output results. After a plurality of high-quality prompt words are generated, how to quickly and conveniently share the prompt words, so that more people can enjoy better service of a large model, and the method is a valuable work in the field of artificial intelligence.
Description
Technical Field
The invention belongs to the field of large model artificial intelligence, and particularly relates to a method and a device for prompting word sharing.
Background
The artificial intelligence large model inputs information which is prompt words (prompt), wherein the prompt words are composed of contents such as purposes, reference information and the like. For the multi-mode artificial intelligence large model, the generalized prompt word also comprises multi-mode information types such as pictures, voice, video and the like. Because of the characteristics of the large model, the prompt words with different qualities can lead to great difference of output results, and even promote special working posts for writing the prompt words.
After a plurality of high-quality prompt words are generated, how to quickly and conveniently share the prompt words, so that more people can enjoy better service of a large model, and the method is a valuable work in the field of artificial intelligence.
Disclosure of Invention
The invention aims to provide a general and convenient method and device for sharing artificial intelligent prompt words.
In order to achieve the above functions, there are first required a "sharer a" (hereinafter referred to as a), a "sharer B" (hereinafter referred to as B), and a "generalized prompt word set P" (hereinafter referred to as P or prompt word P) of good quality, where P is composed of one or more prompt words. For example, P may be a word prompt, or may be a word prompt and a picture prompt, or may be multiple prompts in a dialogue order. And A shares the prompt word P to B, wherein the sharing mode comprises but is not limited to two-dimensional codes, links and pushing. The hint word P is received by a specified "artificial intelligence large model M" (hereinafter referred to as M or model M), and after the large model is processed in sequence, the result of the model is given. M is specified by B or pre-negotiated.
The method can enable the receiving sharer B to automatically use the high-quality prompt word P, so that the manual copying of all or part of the prompt word P by the receiving sharer B is avoided. In order to achieve sufficient usability and convenience, the sharing process should not have explicit downloading or installation procedures.
The specific implementation method comprises the following steps:
s1, sharing a generalized prompt word set P to a receiving sharer B by a sharer A, wherein the generalized prompt word set P consists of one or more prompt words;
s2, receiving a generalized prompt word set P by a designated artificial intelligent large model M;
s3, the designated artificial intelligence large model M processes the prompt word P, and a model result is given.
S4 sharing mode includes but is not limited to two-dimensional code, link and push.
The above steps can simply realize the intention of the invention, but in order to be more general, the shared information needs some 'hint word description information D' (hereinafter referred to as D) besides generalized hint words P, P is used for large models, and D is used for non-large models. The content of D includes, but is not limited to, a specified model, an author, a price, an evaluation, judgment conditions related to the prompt word and a control instruction. The judgment condition refers to the operation condition of judging certain prompt words, for example, a prompt word P related to a physical examination report, and if B is male, the prompt words related to gynecological items in P are not used. The control instruction refers to an instruction of interaction in the environment related to the B, for example, the control instruction in the D may require that the device used by the B take a photo and use the photo as a prompt word. The implementation method comprises the following steps:
s5, the sharer A shares the information of the receiving sharer B, and the content of the prompting word description information D comprises a specified model, an author, a price, an evaluation and a judgment condition and a control instruction related to the prompting word, wherein the content of the prompting word description information D is not limited to the specified model.
In order to increase the applicability of the method to a greater extent, the prompt words in P can be personalized, and the personalization can be divided into personalization before sharing and personalization after sharing. Individuation before sharing, namely, A carries out individuation modification on P which is supposed to be shared according to various information; after sharing, personalization means that after receiving the sharing, the B performs personalized modification on P according to various information before processing the P by the M. The various information includes, but is not limited to, the following information, information of a, information of B, environmental information. Personalized modification refers to modification, deletion and addition of the prompt words in P. For example, in P shared by a to B, the personal privacy information identification card number containing a may be replaced by "modifying" the related prompt word in "personalization before sharing". For example, P shared by a and B is a prompt word set related to analysis of physical examination report, wherein the prompt word set includes analysis of male specialty and analysis of female specialty, and B wants to analyze its physical examination report, if B is male, the prompt word related to female specialty in P needs to be deleted. The deletion of the hint word may have the same effect as the aforementioned "judgment condition".
The addition of the prompt word is slightly complicated, and the prompt word is required to be combined with a large model (a model other than the model M specified above), and a new prompt word is formed by inquiring a plan of the next step of the large model and added into P.
The implementation method for personalizing P is as follows:
s6, individuation of the prompt word set P is divided into individuation before sharing and individuation after sharing.
S7, personalized modification of the prompting word set P comprises modification of a prompting word and deletion of the prompting word.
S8, personalized modification of the prompting word set P' comprises adding a certain prompting word.
For the results generated by M, this is typically used by B, but access to the generated results may also be specified to facilitate collaboration by multiple persons. For example, the results generated by M may be specified for return to the A use or any third party use. This process typically requires B consent.
S9 for the results given by the "artificial intelligence big model M", a user may be specified, including but not limited to "sharer a", "sharer B", or a third party.
The invention relates to a device for applying artificial intelligence, which comprises the following modules:
the D1 storage module is in charge of storing a prompt word set PA of a sharer A, a prompt word set PB received by the sharer B and an address of a designated artificial intelligent large model M, and the returned result is processed by the large model, and an intermediate address is generated during sharing;
and the D2 forwarding module forwards the PA in the D1 to the PB in the D1 for storage, forwards the PB to the M request big model processing in the D1, and retrieves and stores the result after the M processing in the D1.
The optional modules will also include:
d1 will also contain "hint word description information D" for PA and PB.
And D3, modifying, deleting and adding the PA and the PB in the D1 according to requirements.
As described above, the method and the device of the invention have the following beneficial effects:
1. the labor achievements corresponding to the high-quality prompts can be shared quickly;
2. quick landing beneficial to large model application;
3. the method is favorable for the formation and prosperity of large model markets;
4. the cost of artificial intelligence implementation is substantially reduced.
Drawings
Fig. 1 shows a basic logic diagram of the method.
Description of the embodiments
2 examples are listed, namely examples of the e-commerce industry and legal industry, respectively, to illustrate the broad application prospect of the invention.
An example one is as follows.
The sharer A hopes to help the sharer B to finish commodity return of the E-commerce, the prompt word set P comprises 1 prompt word, the mobile phone purchased in the E-commerce yesterday is returned, and the sharer A also comprises corresponding prompt word description information D which is a control instruction and is used for interacting with the running environment of the E-commerce B when the large model is required to execute operation, and a camera is opened to take a picture of the returned commodity. And A, forming a link by the prompt word, and sharing the link to B through social software. And B, after receiving the link, clicking to jump to the large model appointed by both sides, automatically filling the prompting words corresponding to the link into the large model, and calling a camera in the execution process, so that the B finishes shooting of the photo. Finally, the large model helps B to finish commodity return according to the prompt words of A.
Example two is as follows.
In a law house, there are several different working posts to perform collaborative work, and in this example, three working post types are included, namely lawyers, clients communicate and finance.
The lawyer's working content is to communicate with clients in earlier stage, and determine information such as case content, difficulty, price, etc.
The customer communication work content is to pay and communicate with the customer.
Financial work is to determine the receipt of money.
When the lawyer completes the early-stage communication, a detailed description of the case is formed. The lawyer is taken as a sharer A, the case information is taken as one of prompt words, and the requirement of the law-added institute for fee payment is taken as the prompt word, and then the push is formed together and sent to a client communication staff B. And B, after the pushing is opened, the prompt word is operated through the appointed large model, so that a fee result is obtained, and the fee is communicated to the client by the B. Meanwhile, the result obtained by the large model also designates the financial staff C as one of the users to perform effective work collaboration.
Because legal cases often contain sensitive information of the customer, such as property information. The customer usually does not need to know the sensitive information in communication with the B, and the sensitive information contained in the prompt word can be subjected to personalized modification in order to improve the information security. Before sharing the prompt word P to B, the A can carry out personalized modification on the P, delete sensitive information contained in the case, form a new prompt word P', and then share the prompt word P to B.
The customer is overseas and the payment means is also an overseas bank. The original prompt word defaults to an in-house payment mode and generates payment information, which is not in line with the reality of the customer. Therefore, after receiving the prompt word P, the B adds a prompt message, "the client wants to pay by cross-border payment, please prompt the payment process and other related notes, respectively. If the step cannot be completed, please expand the prompt word, and complete the step by step. The large model will give corresponding information based on the prompt and ask itself which ones of the further notes. The large model finds that cross-border payments can generate different tax flows, so the prompt word "detailed description tax flow" can be further generated. This process enables the addition and automatic addition of a cue word in the set of cue words P.
Claims (10)
1. A method for sharing prompt words is characterized in that,
s1, sharing a generalized prompt word set P to a receiving sharer B by a sharer A, wherein the generalized prompt word set P consists of one or more prompt words;
s2, receiving a generalized prompt word set P by a designated artificial intelligent large model M;
s3, the designated artificial intelligence large model M processes the prompt word P, and a model result is given.
2. The method for sharing hint words of claim 1, wherein the S4 sharing manner includes but is not limited to two-dimensional code, link, push.
3. The method of claim 1, wherein S5 "sharer a" shares information to "receiving sharer B", further includes "description of the hint" and "description of the hint" includes, but is not limited to, specified model, author, price, evaluation, judgment condition related to the hint, and control instruction.
4. The method of claim 1, wherein S6 is used for individualizing the "hint word set P" in a manner of pre-sharing individualization and post-sharing individualization.
5. The method of claim 1, wherein the personalized modification of the S7 "set of alert words P" includes modifying an alert word and deleting an alert word.
6. The method of claim 1, wherein the personalized modification of the S8 "set of keywords P" includes adding a keyword.
7. The method of claim 1, wherein S9 specifies a user for the result given by the large artificial intelligence model M, and the user includes, but is not limited to, "sharer a", "sharer B", or a third party.
8. The device for prompting word sharing is characterized by comprising the following modules:
the D1 storage module is in charge of storing a prompt word set PA of a sharer A, a prompt word set PB received by the sharer B and an address of a designated artificial intelligent large model M, and the returned result is processed by the large model, and an intermediate address is generated during sharing;
and the D2 forwarding module forwards the PA in the D1 to the PB in the D1 for storage, forwards the PB to the M request big model processing in the D1, and retrieves and stores the result after the M processing in the D1.
9. The apparatus of claim 8, wherein D1 further comprises "hint word description information D" corresponding to PA and PB.
10. The apparatus for alert word sharing according to claim 8, comprising:
and D3, modifying, deleting and adding the PA and the PB in the D1 according to requirements.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311545314.5A CN117474090A (en) | 2023-11-20 | 2023-11-20 | Prompt word sharing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311545314.5A CN117474090A (en) | 2023-11-20 | 2023-11-20 | Prompt word sharing method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117474090A true CN117474090A (en) | 2024-01-30 |
Family
ID=89636171
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311545314.5A Pending CN117474090A (en) | 2023-11-20 | 2023-11-20 | Prompt word sharing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117474090A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103678963A (en) * | 2012-09-11 | 2014-03-26 | 腾讯科技(深圳)有限公司 | User interaction method, device and system based on graph |
CN108120446A (en) * | 2017-12-21 | 2018-06-05 | 伍军 | Shared based on dynamic web page with the position of Quick Response Code and navigation methods and systems |
CN111144986A (en) * | 2019-12-25 | 2020-05-12 | 清华大学 | Commodity recommendation method and device for social e-commerce website based on sharing behavior |
CN116644145A (en) * | 2023-07-26 | 2023-08-25 | 北京仁科互动网络技术有限公司 | Session data processing method, device, equipment and storage medium |
CN116860935A (en) * | 2023-07-05 | 2023-10-10 | 康键信息技术(深圳)有限公司 | Content management method, device, equipment and medium based on prompt word question-answer interaction |
CN116881429A (en) * | 2023-09-07 | 2023-10-13 | 四川蜀天信息技术有限公司 | Multi-tenant-based dialogue model interaction method, device and storage medium |
CN116993861A (en) * | 2023-08-21 | 2023-11-03 | 北京百度网讯科技有限公司 | Pattern generation method and device and electronic equipment |
CN116992081A (en) * | 2023-08-29 | 2023-11-03 | 中国建设银行股份有限公司 | Page form data processing method and device and user terminal |
-
2023
- 2023-11-20 CN CN202311545314.5A patent/CN117474090A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103678963A (en) * | 2012-09-11 | 2014-03-26 | 腾讯科技(深圳)有限公司 | User interaction method, device and system based on graph |
CN108120446A (en) * | 2017-12-21 | 2018-06-05 | 伍军 | Shared based on dynamic web page with the position of Quick Response Code and navigation methods and systems |
CN111144986A (en) * | 2019-12-25 | 2020-05-12 | 清华大学 | Commodity recommendation method and device for social e-commerce website based on sharing behavior |
CN116860935A (en) * | 2023-07-05 | 2023-10-10 | 康键信息技术(深圳)有限公司 | Content management method, device, equipment and medium based on prompt word question-answer interaction |
CN116644145A (en) * | 2023-07-26 | 2023-08-25 | 北京仁科互动网络技术有限公司 | Session data processing method, device, equipment and storage medium |
CN116993861A (en) * | 2023-08-21 | 2023-11-03 | 北京百度网讯科技有限公司 | Pattern generation method and device and electronic equipment |
CN116992081A (en) * | 2023-08-29 | 2023-11-03 | 中国建设银行股份有限公司 | Page form data processing method and device and user terminal |
CN116881429A (en) * | 2023-09-07 | 2023-10-13 | 四川蜀天信息技术有限公司 | Multi-tenant-based dialogue model interaction method, device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220108311A1 (en) | Application for creating real time smart contracts | |
CN110111193B (en) | Data processing method and device | |
CN103700003A (en) | House online direct renting method and system based on wish conformity matching | |
CN111476637B (en) | Commodity information management method, host platform and commodity information management component | |
WO2022134836A1 (en) | Article value evaluation method and apparatus, and computer device and storage medium | |
CN109345190A (en) | A kind of data processing method and device | |
CN106296154B (en) | Transaction processing method and system | |
CN111242104A (en) | Service calling method and device | |
CN111552793A (en) | Voice outbound method, device, terminal equipment and medium based on outbound robot | |
Panduwinata et al. | BPMN approach in blockchain with hyperledger composer and smart contract: Reservation-based parking system | |
CN106302368A (en) | Transaction methods and device | |
US10699249B2 (en) | Point-of-contact database information integration system | |
US11282174B1 (en) | System and method of providing privacy by blurring images of people in unauthorized photos and videos | |
WO2022094939A1 (en) | Multi-task suspended visual framework application processing method and system based on multi-task operating environment | |
CN112487453A (en) | Data security sharing method and device based on central coordinator | |
CN117474090A (en) | Prompt word sharing method and device | |
Eriksson et al. | MarketSpace’96-An Open Agent-Based Market Infrastructure | |
CN115776548A (en) | Double recording system | |
CN116166514A (en) | Multi-channel data linkage processing method, device, computer equipment and storage medium | |
CN110046233A (en) | Problem distributing method and device | |
CN115239188A (en) | Business handling method and device, electronic equipment and storage medium | |
TWM610290U (en) | A multi task floating visual framework application processing system based on multi task environment | |
CN111832055A (en) | Authority verification system and method | |
US20230153778A1 (en) | System and method for transferring data during a payment process | |
Басшыкызы | SMART WARDROBE-MOBILE APP |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |