Data acquisition method, reasoning method and device for enhancing large model reasoning

Abstract

The invention discloses a data acquisition method and device for enhancing large model reasoning and a reasoning method and device for enhancing large model reasoning. The data acquisition method comprises the following steps: acquiring a plurality of associated original texts from different retrieval sources according to a target demand, splitting each original text into a plurality of text blocks, constructing a conflict graph based on all the text blocks, and determining the source confidence of the retrieval source of each original text; determining the conflict credibility of each node according to the conflict graph and the confidence of all sources; determining the life cycle state credibility of each node according to requirements, determining the interaction feedback credibility of each node according to the interaction metadata and the source credibility corresponding to each node, and further determining the comprehensive analysis credibility of the single node; and taking the original texts corresponding to all the nodes of which the comprehensive analysis credibility is greater than a preset credibility threshold as target data for enhancing large model reasoning. According to the method, high-quality and high-reliability data can be provided, and the performance of large model reasoning is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence information retrieval enhancement, and particularly relates to a data acquisition method and an inference method for enhancing large model inference. BACKGROUND

[0002] Traditional retrieval augmentation generation (RAG) systems usually highly rely on vector similarity-based recall and ranking mechanisms when processing information. This method performs well when facing structured and static knowledge bases. However, with the increasing complexity of the information ecosystem, especially when fast-iterating technical documents, multi-source heterogeneous knowledge bases, and highly diverse community-generated content are introduced, the existing technology faces severe challenges. Vector similarity is difficult to effectively distinguish the authority of information sources and cannot handle conflicts between different versions of content. It is also prone to misuse of semantic fragments that are only valid in specific contexts.

[0003] In addition, valuable verification evidence such as likes and endorsements in community content is often difficult to structure and utilize, leading to the amplification of false information by large language models (LLM) and the generation of inaccurate or misleading results.

[0004] To address the above challenges, some existing technologies attempt to directly calculate a credibility score for documents or paragraphs. However, these methods are essentially a simple linear combination of various weights and lack the ability to infer deep structures and complex relationships behind information. This linear combination strategy can easily be "evaded" by carefully constructed content, i.e., by artificially manipulating text features or metadata to obtain a higher credibility score, while the content itself may not be reliable. More importantly, this shallow evaluation mechanism is insufficient to effectively support various conflicts that exist in a complex knowledge ecosystem and cannot deeply understand the relativity and context dependence of information, making it difficult to effectively resolve conflicts. SUMMARY

[0005] Therefore, the embodiments of the present disclosure provide a data acquisition method and an inference method for enhancing large model inference, which can solve the limitations of traditional RAG methods in handling dynamic, multi-source, and conflicting information, as well as the shortcomings of current credibility evaluation methods, and can achieve credibility enhancement for multi-source conflict resolution.

[0006] In a first aspect, the embodiments of the present disclosure provide a data acquisition method for enhancing large model inference, comprising: acquiring a plurality of original texts associated with different retrieval sources according to target requirements; splitting each of the original texts into a plurality of text blocks; construct a conflict graph based on all the text blocks; the conflict graph comprises relationship edges between each two of the original texts and intensity weights thereof, and each node in the conflict graph is an original text; determine a source confidence of a retrieval source of each original text; determine a conflict credibility of each node according to the conflict graph and all the source confidences; determine a life cycle state credibility of each node according to the target requirement; determine an interaction feedback credibility of each node according to corresponding interaction metadata of each node and the source confidence; determine a comprehensive analysis credibility of each node according to the conflict credibility, the life cycle state credibility and the interaction feedback credibility; take the original texts corresponding to all the nodes with a comprehensive analysis credibility greater than a preset credibility threshold as target data for enhancing model reasoning.

[0007] In a second aspect, the embodiments of the present disclosure further provide a method for enhancing RAG model reasoning, comprising: obtain a plurality of original texts associated from different retrieval sources according to a target requirement; divide each of the original texts into a plurality of text blocks; construct a conflict graph based on all the text blocks; the conflict graph comprises relationship edges between each two of the original texts and intensity weights thereof, and each node in the conflict graph is an original text; determine a source confidence of a retrieval source of each original text; determine a conflict credibility of each node according to the conflict graph and all the source confidences; determine a life cycle state credibility of each node according to the target requirement; determine an interaction feedback credibility of each node according to corresponding interaction metadata of each node and the source confidence; determine a comprehensive analysis credibility of each node according to the conflict credibility, the life cycle state credibility and the interaction feedback credibility; take the original texts corresponding to all the nodes with a comprehensive analysis credibility greater than a preset credibility threshold as target data; output the target data as a reasoning result through a large model.

[0008] In a third aspect, the embodiments of the present disclosure further provide a computer device, adopting the following technical solution: the computer device comprises: at least one processor; and a memory communicatively connected with the at least one processor; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the data acquisition method for enhancing large model inference or the large model inference method of enhanced RAG as described in any one of the above.

[0009] In a fourth aspect, the embodiments of the present disclosure further provide a computer readable storage medium storing computer instructions for causing a computer to execute the data acquisition method for enhancing large model inference or the large model inference method of enhanced RAG as described in any one of the above.

[0010] In a fifth aspect, the embodiments of the present disclosure further provide a computer program product comprising computer programs / instructions which, when executed by a processor, implement the steps of the method as described in any one of the above.

[0011] The data acquisition method for enhancing large model reasoning disclosed in the application firstly acquires a plurality of original texts associated according to target requirements from different retrieval sources, ensures that the corpus acquired is developed around the target requirements rather than random grabbing, and can cover more extensive information perspectives and knowledge of different sources by acquiring from different retrieval sources, thereby avoiding information being limited to a single dimension; then each original text is split into a plurality of text blocks, so that analysis can be in-depth to the local level of the text; then a conflict graph is constructed based on all the text blocks, which systematically quantifies the mutual relationship between different original texts, especially potential conflicts; then the source confidence of the retrieval source of each original text is determined, which introduces trust degree evaluation of the information source itself; then according to the conflict graph and all source confidences, the life cycle state confidence of each node is determined according to the target requirements, the interaction feedback confidence of each node is determined according to the interaction metadata corresponding to each node and the source confidence, and the comprehensive analysis confidence of each node is determined according to the conflict confidence, the life cycle state confidence and the interaction feedback confidence; finally, the original texts corresponding to all nodes with a comprehensive analysis confidence greater than a preset confidence threshold are taken as a corpus for enhancing large model reasoning, wherein the interaction feedback provides a qualitative evaluation from the perspective of user feedback, makes up for the deficiency of pure technical analysis, and can identify outdated, invalid or updated information by evaluating the life cycle state of the text, thereby ensuring that the model uses the latest and most effective information; the conflict confidence of each node is determined, which can more accurately evaluate the conflict confidence of an original text in the whole information set by combining the structured conflict relationship between texts (conflict graph) and the reliability of the source itself (source confidence); the three independent confidence dimensions (conflict, life cycle and interaction feedback) are organically combined to form a more comprehensive and robust “comprehensive analysis confidence”, which can more accurately reflect the real value and reliability of the text by fusing multiple evaluation dimensions, thereby improving the accuracy and robustness of evaluation, and finally through strict threshold screening, each original text entering the corpus has undergone multi-dimensional and fine-grained confidence testing. When the large model uses the corpus that has been screened in such a strict manner for reasoning, the authenticity, accuracy and consistency of the output results will be greatly improved, effectively reducing the generation of illusions and false information, providing reliable information increments for the large model, enabling it to generate more persuasive and easier-to-rely-on answers, thereby reducing the potential risks of the model in actual application, effectively solving the limitations of traditional RAG methods in handling dynamic, multi-source and conflict information, and the deficiencies of current existing confidence evaluation methods and other problems.

[0012] The above description is only a summary of the technical solutions of the present disclosure. In order to more clearly understand the technical means of the present disclosure, the contents of the specification can be implemented, and in order to make the above and other purposes, characteristics and advantages of the present disclosure more obvious and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0014] Figure 1 The flowchart of the data acquisition method for enhancing large model inference provided by the embodiments of the present disclosure.

[0015] Figure 2 The flowchart of the method for acquiring a plurality of original texts associated according to target requirements from different search sources provided by the embodiments of the present disclosure.

[0016] Figure 3 The flowchart of the method for constructing a conflict graph based on all text blocks provided by the embodiments of the present disclosure.

[0017] Figure 4 The flowchart of the method for determining the conflict credibility of each node provided by the embodiments of the present disclosure.

[0018] Figure 5 The flowchart of the method for acquiring the life cycle state credibility of each node provided by the embodiments of the present disclosure.

[0019] Figure 6 The flowchart of the method for acquiring the interaction feedback credibility of each node provided by the embodiments of the present disclosure.

[0020] Figure 7 The structural diagram of a computer device provided by the embodiments of the present disclosure. DETAILED DESCRIPTION

[0021] The embodiments of the present disclosure will be described in detail below with reference to the drawings.

[0022] It should be apparent that the following describes embodiments of the present disclosure by way of specific examples, and that one skilled in the art could readily derive other advantages and / or effects from this description. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, and are not all-inclusive. The present disclosure can also be implemented or applied by other different specific embodiments, and the details in the description can be modified or changed based on different views and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by one of ordinary skill in the art without creative labor are within the scope of protection of the present disclosure.

[0023] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be implemented in a wide variety of forms and that any particular structure and / or function described herein is merely illustrative. Based on the teachings herein one skilled in the art will appreciate that one or more aspects described herein can be implemented independently of any other aspects and that an aspect can be implemented both as any number of software, firmware, and / or hardware structures. For example, an aspect can be implemented as a single structure on an integrated circuit or a structure implemented in multiple structures such as software running on a processor or software run on a hardware co-processor. An aspect can also be implemented as software stored on a medium that when run can implement an aspect. An aspect can be implemented as hardware that can be one time programmable, multi-time programmable, and / or configurable, such as with reconfigurable and / or configurable logic. An aspect can also be implemented as firmware that can be one time programmable, multi-time programmable, and / or configurable, such as with reconfigurable and / or configurable logic. An aspect can also be implemented as hardware, where the hardware can be one time programmable, multi-time programmable, and / or configurable, such as with reconfigurable and / or configurable logic that when configured implements an aspect. An aspect can also be implemented as a combination of software / circuitry that when executed / operational implements the aspect. An aspect can also be implemented as software, hardware, firmware and / or circuitry in different embodiments.

[0024] It should also be noted that the drawings included in the following description are for illustrative purposes only and that they are not drawn to scale. Certain aspects can be utilized in implementations that are not depicted in the drawings. It should also be noted that the drawings are not to be considered as defining the scope of the disclosure.

[0025] Further, in the following description, numerous specific details are provided for a thorough understanding of examples. One skilled in the relevant art will recognize, however, that the aspects described herein can be practiced without one or more of these specific details.

[0026] Referring to Figure 1 The present disclosure discloses a data acquisition method for enhancing large model inference, which can realize credibility enhancement of multi-source conflict resolution. The method specifically comprises: S100, acquiring a plurality of original texts associated from different retrieval sources according to a target requirement.

[0027] S200, splitting each original text into a plurality of text blocks.

[0028] S300, constructing a conflict graph based on all text blocks.

[0029] wherein the conflict graph comprises relationship edges between each pair of original texts and their intensity weights, and each node in the conflict graph is an original text.

[0030] S400, determine the source confidence of the retrieval source of each original text.

[0031] S500, determine the conflict credibility of each node according to the conflict graph and all source confidences.

[0032] S600, determine the life cycle state credibility of each node according to the target requirement.

[0033] S700, determine the interaction feedback credibility of each node according to the interaction metadata corresponding to each node and the source confidence.

[0034] S800, determine the comprehensive analysis credibility of each node according to the conflict credibility, the life cycle state credibility, and the interaction feedback credibility.

[0035] S900, take the original text corresponding to each node with a comprehensive analysis credibility greater than a preset credibility threshold as the corpus (i.e., target data) for enhancing the reasoning of the large model.

[0036] Referring to Figure 2 For the method of S100 "obtaining several original texts associated with the target requirement from different retrieval sources", the method specifically comprises: S110, determine the intent type of the target requirement.

[0037] Specifically, a lightweight classification model (such as BERT-Tiny or a rule-based keyword matching) can be used to classify the intent of the user Query, and the corresponding intent type is obtained.

[0038] wherein the intent type is any one of API usage type, version change type, troubleshooting type, and best practice type.

[0039] API usage type refers to a specific type of information requirement or user query category, which can include: asking function parameters, return values, and configuration items; the feature words corresponding to the API usage type include: config, parameter, how to use, argument, etc., that is, when one or more of these words appear in the Query input by the user, it can be determined that the Query belongs to the "API usage type".

[0040] The version change class can include: asking new features, deprecated functions, migration guide; the feature words corresponding to this class: deprecated, new in v5, migration, difference, etc., that is, when one or more of these words appear in the Query (query) input by a user, it can be determined that the Query belongs to the "version change class".

[0041] The troubleshooting class includes: asking for error information, Bug repair; the feature words corresponding to this class include: error, exception, fail to start, bug, etc., that is, when one or more of these words appear in the Query (query) input by a user, it can be determined that the Query belongs to the "troubleshooting class".

[0042] The best practice class includes: asking about architecture design, performance optimization; the feature words corresponding to this class include: bestpractice, optimization, architecture, etc., that is, when one or more of these words appear in the Query (query) input by a user, it can be determined that the Query belongs to the "best practice class".

[0043] S120, determining the target source list corresponding to the intent type according to the preset routing mapping table.

[0044] Table 1: Preset routing mapping table Referring to Table 1, the preferred search source refers to the authoritative core source, the secondary search source refers to the code / engineering source, and the excluded / weighted source refers to the community / experience source; the target source list includes the summary information of the preferred search source, the secondary search source, and the excluded / weighted source.

[0045] S130, identifying the key technical entity in the target demand.

[0046] Specifically, a pre-trained NER model can be used, which is specially trained for the technical document field, and then the key technical entity in the target demand is identified based on the trained NER model.

[0047] S140, obtaining a plurality of original texts associated with the key technical entity from the corresponding search source according to the target source list.

[0048] Further, if the entity is an Open Source Project (such as React): the GitHub repository address of the project is automatically added; if the entity is a Cloud Service (such as AWS): the official SDK document address of the service is automatically added.

[0049] For example, in Example 1 (Neo4j configuration), the user Query is: "How to configure n10s plugin in Neo4j version 5.26?" The intent recognition includes: keyword configuration, determined as API usage type; keyword 5.26, determined to contain version constraints.

[0050] The source routing decision includes: first preference for Neo4j official documentation (Tier 1), especially Operations Manual; second preference for GitHub neo4j-contrib / neosemantics repository (Tier 2) for verifying specific version behaviors of the plugin. Further supplementary information includes: Neo4j community forum (Tier 3) for finding the latest user pitfalls records. Exclusion: CSDN / Medium blogs before 2022 (because the Query specifies the very new version 5.26, old blogs are high-risk noise).

[0051] In this example, through the three-layer strategy of "intent recognition, route mapping, entity expansion", the range of data sources most likely to contain correct answers can be accurately circled, which not only improves the retrieval efficiency, but also reduces the probability of low-quality information (such as outdated blogs) entering the system from the source.

[0052] In this example, S100 ensures that the corpus obtained is around the target demand, rather than random scraping; by obtaining from different retrieval sources, it can cover a wider range of information perspectives and knowledge from different sources, avoiding information limited to a single dimension, laying a foundation for subsequent more detailed analysis, emphasizing the original text of the association, indicating that preliminary information screening has been performed, excluding texts unrelated to the target demand, which can effectively improve the efficiency and effectiveness of subsequent processing.

[0053] For S200 "split each original text into several text blocks", specifically includes: S210, perform parsing processing on each original text to extract pure text segments containing text content, remove embedded format markers and non-text elements, and obtain target text.

[0054] S220, divide each target text into several discrete text blocks.

[0055] Specifically, each target text is disassembled into independent, verifiable, core, and fact-based information claims, i.e., divided into several discrete text blocks. Through this step, the transition from "unstructured documents" to "structured knowledge points" can be achieved for subsequent comparison.

[0056] Assume we retrieve the following two raw document snippets in this step, which discuss the same technical point (i.e., correspond to the same entity) but have conflicting content. Perform parsing on each raw text to extract the plain text fragment containing the text content, remove embedded format markers and non-text elements, and obtain the target text.

[0057] Target Text A (from Neo4j 4.x official documentation in 2021); Text: "To configure the n10s extension in Neo4j, you must add the following line to your neo4j.conf file: dbms.unmanaged_extension_classes=n10s.endpoint= / rdf. After saving the file, a server restart is required." Meta: {id: "doc_A", ver: "4.2", source: "official"}.

[0058] Target Text B (from Neo4j 5.x migration guide in 2024); Text: "In Neo4j 5, the dbms.unmanaged_extension_classes setting has been removed. You should now register n10s using server.dynamic_setting.allow_list or via the Cypher procedure CALL n10s.graph.config(...)." Meta: {id: "doc_B", ver: "5.26", source: "official"}.

[0059] Then the LLM can be called to disassemble the above two texts into independent, atomic claim nodes, i.e., to split each target text into several discrete text blocks.

[0060] In the process of calling the LLM, the instructions issued can include "Please disassemble the following text into several independent atomic facts, with the requirements 1) each claim must be a complete sentence containing the subject; 2) the version limit, configuration parameter name, and operation condition in the original text must be preserved; 3) the format is JSON List".

[0061] The segmentation results of Target Text A include: Prop A1 : "In Neo4j 4.x, configuring n10s requires editing theneo4j.conf file."Prop A2 : "In Neo4j 4.x, the configuration parameter isdbms.unmanaged_extension_classes."Prop A3 : "Configuring n10s in Neo4j 4.xrequires a server restart." The segmentation results of the target text B include: Prop B1 : "In Neo4j 5, the dbms.unmanaged_extension_classes setting isremoved."Prop B2 : "In Neo4j 5, n10s should be registered using server.dynamic_setting.allow_list."Prop B3 : "In Neo4j 5, n10s can be configured via Cypherprocedure CALL n10s.graph.config."Through this step, 6 independent proposition nodes are obtained, i.e. 6 text blocks, and at this time the obtained information is a mixture of correct and incorrect information.

[0062] In this embodiment, S200 splits the original text into smaller text blocks, enabling analysis to delve into the local level of the text, which is crucial for identifying subtle information, potential conflicts or different arguments within the text, and lays the foundation for constructing a detailed conflict map; this embodiment allows independent evaluation of each part of the text, making it possible to more effectively mine and utilize scattered but valuable information in the text.

[0063] Referring to Figure 3 , the method of S300 "constructing a conflict map based on all text blocks" specifically includes: S310, obtaining relationship comparison results between two original texts based on all text blocks.

[0064] The relationship comparison results are contradiction, competition or juxtaposition.

[0065] In this embodiment, each original text corresponds to several text blocks constituting an information claim. A pre-trained general NLI model (such as BER, RoBERTa, DeBERTa, etc.) can be used to determine the relationship between two information claims, or a specific technical field NLI model can be used to determine the relationship between two information claims, and the relationship comparison results between each pair of original texts are obtained.

[0066] If it is determined that the relationship between the two information claims is mutual exclusion or negation, it is determined that the relationship comparison result is contradictory; if it is determined that the relationship between the two information claims is a solution to the same problem but mutually exclusive, it is determined that the relationship comparison result is competitive; if it is determined that the two information claims are both feasible paths to complete a certain task and have no direct conflict or substitution relationship with each other, it is determined that the relationship comparison result is parallel.

[0067] It should be noted that this application obtains a plurality of original texts associated from different search sources based on target requirements, and therefore the entities of the information claims corresponding to all original texts are the same.

[0068] S320, mapping the information claim corresponding to each original text into a node in the graph database.

[0069] The graph database can be Neo4j, ArangoDB, Amazon Neptune, etc.

[0070] S330, generating relationship edges and their strength weights between each pair of nodes according to the relationship comparison results.

[0071] The relationship edges include negative edges, weak negative edges or positive edges, and the strength weight corresponding to the negative edge is less than the strength weight corresponding to the weak negative edge, and the strength weight corresponding to the weak negative edge is less than the strength weight corresponding to the positive edge.

[0072] The relationship edges (i.e. edge types) and strength weights generated in this step are mapped to the relationship between nodes, i.e. in the embodiment, all generated relationships (edges) and nodes are batch imported into the graph database, and the conflict graph is obtained.

[0073] Further, the strength weight corresponding to the negative edge is preferably -1.0, indicating strong conflict and mutual negation. The strength weight corresponding to the weak negative edge is preferably -0.8, indicating that there is conflict or mutual exclusion, but it may not be absolutely negated, and there is a degree of weakening relationship. The strength weight corresponding to the positive edge is preferably 1.0 (indicating support, consistency, complementarity) or 0 (indicating no direct relationship, no edge added).

[0074] For example, comparison group 1: Prop A2 vs Prop B1, the NLI inference result is: Contradiction. Entity: dbms.unmanaged_extension_classes. Operation: Create Prop A2 Node (if not exist) and Prop B1 Node (if not exist). In Prop A2 and Prop B1 Establish a CONTRADICTS relationship between the two nodes, and set the strength weight of the relationship: -1.0.

[0075] Comparison group 2: Prop A2 vs Prop B2 , the NLI inference result is: Competition. Entity: dbms.unmanaged_extension_classes (Prop A2 ) vs server.dynamic_setting (PProp B2 ). Note: The entities here are different, but the system needs to judge that they are in competition under the intent of "how to configure some similar function of Neo4j". Prop A2 mentions a specific parameter, Prop B2 mentions a list of allowed settings. Operation: In Prop A2 and Prop B2 Establish a COMPETES_WITH relationship between the two nodes. Set the strength weight of the relationship: -0.8.

[0076] Comparison group 3: Prop B2 vs Prop B3 , the NLI inference result is: Parallel. Entity: server.dynamic_setting (Prop B2 ) vs Cypher CALL (Prop B3 ). Context: Both can be recommended ways to implement a certain function in Neo4j 5.x version. Operation (assuming strategy A, establish positive edges): In Prop B2 and Prop B3 Establish a SUPPORTS relationship between the two nodes. Set the strength weight of the relationship: +1.0.

[0077] Operation (assuming strategy B, no edge): Do not perform any operation, the two nodes are not directly connected in the graph at present.

[0078] Further, we can draw the JSON into a graph: such as Prop B1(5.x) A red arrow (attack / negation) is fired, pointing to Prop A2 (4.x) Prop B2 (5.x) Stand next to Prop B1 As a new alternative, this structure clearly tells the following algorithm: Prop A2 is the attacked object, whose credibility should be reduced; and Prop B1 is the initiator of the attack (and from the new version), whose credibility should be high. This is the basis for "inference of credibility from structure".

[0079] The implementation of constructing the conflict graph in this application is a multi-stage process involving information extraction, NLI inference, relationship mapping, and graph database operations. The key is how to effectively train the NLI model to capture the unique conflicts and competitive relationships in the technical field, and how to design a flexible representation to store these relationships and their strengths.

[0080] Further, when assigning weights, the reliability of the information source can also be considered. For example, claims from official documents should be given higher weight than those from personal blogs. This can be achieved by introducing a "reputation score" outside the NLI model to adjust the final weight.

[0081] In this embodiment, S300 quantitatively identifies the mutual relationships between different original texts, especially potential conflicts, for the first time. The relationship edges and their strength weights can identify the contradictions hidden in the vast amount of information, which is crucial for handling dynamic, multi-source, and conflicting information, avoiding reasoning errors caused by directly adopting mutually contradictory information. By treating original texts as nodes and relationships as edges, a structured way to understand the complex associations between information is provided, facilitating subsequent quantitative analysis.

[0082] For S400 "determine the source confidence of each original text's search source", the official source confidence is greater than the community source confidence.

[0083] Specifically, the old official source confidence is less than the new official statement / announcement / blog source confidence, the ordinary technical blog / forum source confidence is less than the old official source confidence, and the well-known community platform source confidence is less than the new official statement / announcement / blog source confidence and greater than the ordinary technical blog / forum source confidence.

[0084] Specifically, assume there are 4 typical nodes, node V1 (old official), content: "Configure n10s requires modifying dbms.unmanaged_extension_classes." Source: Neo4j v4.2 official document; source confidence: 0.9 (official documents usually have high initial scores).

[0085] Node V2 (new official statement / announcement / blog), content: "The dbms.unmanaged_extension_classes configuration item has been removed." Source: Neo4j v5.26 migration guide; source confidence: 0.95.

[0086] Node V3 (new official document), content: "Please use server.dynamic_setting for configuration." Source: Neo4j v5.26 operation manual; source confidence: 1.

[0087] Node V4 (ordinary technical blog / forum), content: "No need to configure, just copy the jar package and restart." Source: A personal blog (2020); source confidence: 0.4 (personal blogs have low initial scores).

[0088] In this embodiment, S400 introduces trust assessment of the information source itself, different search sources (e.g., official documents, news reports, personal blogs, academic papers, etc.) have different credibility, and this step can quantify this trust, providing an important basis for subsequent comprehensive assessment of credibility; help to distinguish texts from more reliable sources, so as to prefer information from sources with higher credibility when there is a conflict.

[0089] Reference Figure 4 For the method of S500 "determining the conflict credibility of each node according to the conflict graph and the credibility of all sources", the method for determining the conflict credibility of each node specifically includes: S510, according to the source of each node, configure an initial authority score for each node.

[0090] Specifically, the initial authority score can be preset according to the metadata (Source Meta) when the original text is imported. In a preferred embodiment, when the source of the node is an official document (base Tier 1), it is set to 0.9~0.95; when the source of the node is code / engineering source (i.e., Tier 2), it is set to 0.8; when the source of the node is a community blog, etc. (i.e., Tier 3), it is set to 0.4.

[0091] S520, determine the relationship edges and their strength weights of the node to be analyzed from the conflict graph.

[0092] S530, based on the initial authority score and strength weight of the node corresponding to the relation edge of the node to be analyzed, determine the degree of conflict between the node corresponding to the relation edge of the node to be analyzed and the node to be analyzed.

[0093] S540 obtains the conflict credibility of the node to be analyzed based on all conflict levels and the initial authority score of the node to be analyzed.

[0094] No. The conflict credibility of each node is : .

[0095] in, For the first The initial authority score of each node, The influence coefficient, In order to be with the first The node with the associated relationship The initial authority score of each node, For the first The node and the first Strength weights between nodes For the first The node is the first The degree of conflict of each node (i.e., the node to be analyzed).

[0096] Furthermore, This is used to control the influence strength of neighbor node conflict information with semantic overlap. If A larger value indicates that the credibility of the claim relies more heavily on the "support" or "refutation" of other evidence; if A lower score indicates that the credibility of the claim depends more on its own basic authority score.

[0097] In this embodiment, S500 can more accurately assess the conflict credibility of an original text in the overall information set by combining the structured conflict relationship (conflict map) between texts and the reliability of the source itself (source confidence). If a text conflicts with many other sources with lower credibility, its conflict credibility will decrease. Conversely, if a text itself has high source credibility and is consistent with other high-credibility texts, its conflict credibility will remain at a high level even if there are a few conflicts.

[0098] Reference Figure 5 The method for S600 to "determine the lifecycle state credibility of each node according to target requirements," specifically the method for obtaining the lifecycle state credibility of each node, includes: S610, determine the lifecycle status of each node. The lifecycle status includes the existing version time and the corresponding existing version description features.

[0099] S620: Obtain the current time and current version description features corresponding to the target requirement.

[0100] S630 determines the validity type of each node's status based on the current version description features, current time, and lifecycle status.

[0101] S640, determine the corresponding state weight based on the state validity type.

[0102] In this embodiment, if an existing version description feature is explicitly removed from the current version description feature, the state validity type of the corresponding node is determined to be the version removal class, and the corresponding state weight is determined to be 0.01 to prevent complete discarding, but to make its ranking extremely low, so a very small value is assigned.

[0103] If an existing version description feature is explicitly deprecated in the current version description feature, the state validity type of the corresponding node is determined to be version deprecation removal, and the corresponding state weight is determined to be 0.05.

[0104] If the existing version description features are not explicitly defined in the current version description features, and the existing version time is not later than the current time, the status validity type of the corresponding node is determined to be active, and the corresponding status weight is determined to be 1.0.

[0105] S650 determines the lifecycle state reliability of the node based on the state weight and a preset formula.

[0106] The reliability of the lifecycle state corresponding to each node is: : Where D is the version difference between the current time and the existing version time. These are the state weights.

[0107] Specifically, for example, if the target requirement is "How to configure the n10s plugin in Neo4j 5.26?", the current time corresponding to the target requirement is obtained as follows: (i.e., target version), current version description features.

[0108] The nodes to be evaluated include: 1) Prop A (Old method): "Use dbms.unmanaged_extension_classes." Corresponding source metadata: Neo4j v4.2 documentation (released in 2021). 2) Prop B(New method): "Use server.dynamic_setting." Corresponding source metadata: Neo4j v5.26 documentation (released in 2024). 3) Prop C (Intermediate state): "dbms.unmanaged... is marked as Deprecated, migration is recommended." Corresponding source metadata: Neo4j v4.4 changelog (released in 2022).

[0109] For Prop A (v4.2) Prediction: and It crossed a major version number; prediction rule: crossing a major version usually means a breaking change; inferred status: Removed or High_Risk. For Prop C Prediction for (v4.4): The original text mentions "v4.4 Deprecated". Lifecycle trajectory: Active (v4.0) -> Deprecated (v4.4) -> Removed (v5.0+); Prediction conclusion: In v5.26, this feature should have been completely removed; Inferred status: Removed. Regarding Prop... B (v5.26) Prediction: Inferring state: Active (active / valid). Then, the reliability of the lifecycle state for each node can be determined.

[0110] In this embodiment, S600 can identify outdated, invalid, or updated information by evaluating the lifecycle status of the text, ensuring that the model uses the latest and most effective information, filtering out old information that no longer has practical significance, and preventing the model from reasoning based on outdated knowledge, thereby improving the timeliness and accuracy of the response.

[0111] Reference Figure 6 The S700 method of "determining the credibility of each node's interaction feedback based on the interaction metadata and source confidence of each node" specifically includes the following methods for obtaining the credibility of each node's interaction feedback: S710, obtain the interaction metadata corresponding to each node; among which, the interaction metadata includes the number of approvals and the number of confirmations.

[0112] Recognition includes one or more of the following: likes, quantity, and quantity. The quantity of recognition refers to the total number of recognitions. Confirmation includes one or more of the following: positive feedback, being cited, and being adopted. The quantity of confirmation refers to the total number of confirmations.

[0113] S720 determines the initial energy information of each node based on the source confidence level corresponding to each node.

[0114] In this embodiment, the higher the source confidence, the larger the initial energy information is configured. For example, if the source is an official developer / maintainer reply, the initial energy information is preset to the highest (e.g., 0.9); if the source is a high-reputation platform (e.g., StackOverflow accepted answers), the initial energy information is preset to the second highest (e.g., 0.8); and if the source is a regular community post / personal blog, the initial energy information is preset to a lower level (e.g., 0.4).

[0115] The S730 determines the credibility of each node's interactive feedback based on the number of acknowledgements, confirmations, and initial energy information.

[0116] The credibility of interactive feedback is : .

[0117] in, As the recognition coefficient, The number of people who have given their approval (this is the most basic signal of group approval). To confirm the quantity, To confirm the coefficients, This refers to the initial energy information of the corresponding node. In this embodiment, Greater than This step transforms feedback from the community and other sources into structured information, taking into full account user interaction behavior and further improving credibility.

[0118] In this step, for conduct Logarithmic processing is used to prevent the "Matthew effect" in social data, that is, to prevent a viral post with a few thousand likes from having such a high impact that it overshadows all other evidence. The processing slows down the energy growth as the number of likes increases, ensuring the stability of the system.

[0119] In this step, The corresponding confirmation can be obtained through explicit verification feedback extracted from the comments section using LLM (Large Model) or semantic parsing. For example, the number of replies containing phrases such as "personally tested and effective", "solved my problem", and "confirmed to be true".

[0120] in, The preferred value is 0.5. The preferred value is 0.2, that is The weight (0.5) is significantly higher than (0.2) This is because the system believes that "explicit confirmation with textual feedback" has higher technical reference value than "casual likes". This weighting design ensures that even if the number of likes is not high, as long as multiple experts confirm its validity, the energy field of the knowledge point will be rapidly enhanced. At the same time, through the setting of these two weights, it is possible to identify invalid content that "gains high likes simply because it is funny or has a clickbait title" but lacks technical confirmation.

[0121] The formula is set according to the official documentation ( (High) has not yet been updated, but the community issue (although) Low, but When the new configuration method is pointed out (extremely high), the final energy value of the issue will surpass that of the official documentation, effectively solving the drawbacks of the traditional RAG.

[0122] In this embodiment, the interactive metadata in S700 (such as likes, comments, citations, number of adoptions, etc.) reflects the feedback and recognition the text receives in actual applications; combined with source confidence, it can more comprehensively evaluate the actual value and influence of a text; high interactive feedback usually means that the text has high practicality or is widely accepted, which is an independent and important credibility indicator. Interactive feedback provides a qualitative assessment from the perspective of user feedback, making up for the shortcomings of purely technical analysis.

[0123] The S800 method of "determining the comprehensive analysis credibility of each node based on conflict credibility, lifecycle state credibility, and interaction feedback credibility" specifically includes: weighted summation of conflict credibility, lifecycle state credibility, and interaction feedback credibility to obtain the score corresponding to the comprehensive analysis credibility of each node.

[0124] In S800, the three independent credibility dimensions (conflict, lifecycle, and interaction feedback) are organically combined to form a more comprehensive and robust "comprehensive analysis credibility". Any single credibility indicator has limitations. This multi-dimensional assessment method is more comprehensive and in-depth than a single source confidence or simple verification method. It can more accurately reflect the true value of the text. By integrating multiple assessment dimensions, it can more accurately reflect the true value and reliability of the text, and improve the accuracy and robustness of the assessment.

[0125] The S900's provision of "using the original text corresponding to all nodes whose comprehensive analysis credibility exceeds the preset credibility threshold as the corpus for enhancing large-scale model inference" includes: directly outputting the original text corresponding to nodes whose comprehensive analysis credibility exceeds the preset credibility threshold, and deleting other original texts that do not meet the conditions. Through rigorous threshold screening, it ensures that every original text entering the corpus undergoes multi-dimensional and refined credibility checks. When the large model uses such a rigorously screened corpus for inference, the authenticity, accuracy, and consistency of its output results are greatly improved, effectively reducing the generation of illusions and misinformation. This provides reliable information increments for the large model, enabling it to generate more persuasive and trustworthy answers, thereby reducing the potential risks of the model in practical applications.

[0126] In this embodiment, the introduction of lifecycle state credibility can identify and filter out outdated and no longer valid information, ensuring that the model uses the latest and most relevant knowledge. Source confidence and conflict graphs work together to assess the original reliability of information from different retrieval sources and identify potential contradictions between different sources, thus preventing the model from mistakenly accepting unreliable original text. The conflict graph is one of the core innovations; by analyzing the strength of relationships between pairs of original texts, it can quantify and identify conflicts between texts. This allows the model to handle information conflicts more intelligently, for example, by favoring texts with lower conflict levels or marking conflicting parts, rather than mechanically mixing all information and causing chaotic inference results. Interactive feedback credibility utilizes interactive metadata, effectively introducing social proof or user feedback, which can serve as an effective indicator of text acceptance and actual utility—more grounded in reality than purely technical evaluation.

[0127] Finally, the corpus is filtered by comprehensively analyzing texts whose credibility exceeds a preset credibility threshold. This ensures that only high-quality, highly credible texts are used to enhance model inference, thereby improving the credibility of the model's output from the source. Through this meticulous corpus filtering, the model can obtain higher-quality and more accurate retrieval results when performing RAG operations, significantly reducing the occurrence of illusions and improving the authenticity and consistency of generated answers.

[0128] For fields such as finance, healthcare, law, and news analysis, which are characterized by complex information, high real-time requirements, and frequent information conflicts, this method can provide more reliable corpus support, enabling large models to function more safely and effectively in these fields.

[0129] This application discloses a data acquisition method for enhancing large-scale model inference, aiming to address the numerous challenges encountered by traditional RAG (Retrieval-Augmented Generation) in handling complex information and to compensate for the shortcomings of existing credibility assessment methods. This method provides high-quality, high-credibility corpora, significantly improving the performance of large-scale model inference. It offers a streamlined corpus construction mechanism that automates the processing of massive, heterogeneous data, greatly reducing manual intervention and data preprocessing workload. By intelligently filtering the corpus, the amount of data that the model needs to retrieve and process during inference can be effectively reduced, potentially lowering real-time latency and computational costs, and improving overall engineering efficiency.

[0130] Secondly, this application discloses a method for enhancing large-scale model inference in RAG, including: S10, Based on the target requirements, retrieve several related original texts from different retrieval sources; S20: Each original text is split into several text blocks, and a conflict graph is constructed based on all text blocks.

[0131] The conflict graph includes the relationship edges between each pair of original texts and their strength weights. Each node in the conflict graph is an original text. S30, determine the source confidence of the retrieval source for each original text; S40, Based on the conflict graph and the confidence of all sources, determine the conflict confidence of each node; S50 determines the lifecycle status reliability of each node based on target requirements; S60, based on the interaction metadata and source confidence of each node, determine the credibility of the interaction feedback of each node; S70, based on the credibility of conflict, the credibility of life cycle status, and the credibility of interactive feedback, determine the comprehensive analysis credibility of each node; S80, the original texts corresponding to all nodes whose comprehensive analysis credibility is greater than the preset credibility threshold are used as the target corpus; S90 outputs the target corpus as the reasoning result through a large model.

[0132] It should be noted that the schemes protected by S10-S80 are consistent with the data acquisition method for enhancing large model inference disclosed in the first aspect of this application. For details of the specific implementation scheme, please refer to the detailed content disclosed in the first aspect of this application, so they will not be repeated here.

[0133] The method disclosed in this application enables enhanced credibility-aware retrieval, allowing large models to automatically identify credible claims, accurately separate authoritative from low-quality sources, and proactively resolve version conflicts in scenarios involving knowledge conflicts, multiple versions, and mixed low-quality sources. This effectively prevents misinformation from being amplified and significantly improves factual consistency, interpretability, and robustness.

[0134] Traditional RAG methods have limitations in handling dynamic, multi-source, and conflicting information. This application constructs a complex knowledge ecosystem that can effectively incorporate advanced technologies in real time. This ecosystem not only requires the ability to retrieve relevant content from massive amounts of information, but more importantly, it must be able to understand the source, version, and contextual dependencies of information, and evaluate community consensus and verification clues, i.e., perform deep structural inference. This ensures the accuracy and reliability of the information retrieval and generation process, and ultimately improves the application effect of large-scale language models in complex knowledge environments.

[0135] A computer device according to embodiments of the present disclosure includes a memory and a processor. The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0136] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the computer device to perform desired functions. In one embodiment of this disclosure, the processor is used to execute computer-readable instructions stored in the memory, causing the computer device to perform all or part of the steps of the data acquisition method for enhanced large model inference or the enhanced RAG large model inference method described in the foregoing embodiments of this disclosure.

[0137] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0138] like Figure 7 This is a schematic diagram of a computer device provided for an embodiment of the present disclosure. It illustrates a structural schematic diagram suitable for implementing the computer device in the embodiments of the present disclosure. Figure 7 The computer device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0139] like Figure 7 As shown, a computer device may include a processor (such as a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) or programs loaded from storage devices into random access memory (RAM). The RAM also stores various programs and data required for the operation of the computer device. The processor, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0140] Typically, the following devices can be connected to the I / O interface: input devices, such as sensors or visual information acquisition devices; output devices, such as displays; storage devices, such as magnetic tapes or hard drives; and communication devices. Communication devices allow the computer device to communicate wirelessly or wiredly with other devices (such as edge computing devices) to exchange data. Although Figure 7 A computer apparatus with various devices is shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or included alternatively.

[0141] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from ROM. When the computer program is executed by a processor, all or part of the steps of the data acquisition method for enhanced large model inference or the enhanced RAG large model inference method of the embodiments of this disclosure are performed.

[0142] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0143] A computer-readable storage medium according to embodiments of the present disclosure stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the data acquisition method for enhanced large model inference or the large model inference method for enhanced RAG described in the foregoing embodiments of the present disclosure are performed.

[0144] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0145] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0146] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0147] In this disclosure, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The block diagrams of devices, apparatuses, devices, and systems involved in this disclosure are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as "comprising," "including," "having," etc., are open-ended terms meaning "including but not limited to," and are used interchangeably with them. The terms "or" and "and" as used herein refer to the terms "and / or," and are used interchangeably with them unless the context clearly indicates otherwise. The term "such as" as used herein refers to the phrase "such as but not limited to," and is used interchangeably with it.

[0148] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0149] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0150] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0151] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0152] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A data acquisition method for enhancing large model inference, characterized in that, include: Retrieve several related original texts from different search sources based on the target requirements; Each of the original text entries is split into several text blocks; A conflict graph is constructed based on all the text blocks; the conflict graph includes the relationship edges between each pair of the original texts and their strength weights, and each node in the conflict graph is one of the original texts; Determine the source confidence level of the retrieval source for each of the original texts; Based on the conflict map and all the source confidence scores, determine the conflict confidence score for each node; Determine the lifecycle status credibility of each node based on the stated target requirements; The credibility of the interaction feedback of each node is determined based on the interaction metadata corresponding to each node and the source confidence level. Based on the conflict credibility, the lifecycle state credibility, and the interaction feedback credibility, the comprehensive analysis credibility of each node is determined. The original text corresponding to all nodes whose comprehensive analysis credibility is greater than a preset credibility threshold is used as the target data for enhancing the large model inference.

2. The data acquisition method for enhancing large model inference according to claim 1, characterized in that, The process of retrieving several related original texts from different retrieval sources according to the target requirements includes: Determine the type of intent for the target need; The list of target sources corresponding to the intent type is determined according to the preset routing mapping table; Identify the key technical entities in the target requirements; Based on the target source list, retrieve several original texts associated with the key technology entity from the corresponding retrieval sources; The number of original text entries is not less than the number of sources of the retrieval source; Each original text entry includes at least source information, version information, publication time information, and interaction records.

3. The data acquisition method for enhancing large model inference according to claim 2, characterized in that, The step of splitting each original text into several text blocks includes: Each piece of original text is parsed to extract the plain text fragment containing the text content, remove embedded formatting tags and non-text elements, and obtain the target text; Each target text is divided into several discrete text blocks; Several discrete text blocks include at least version information and interaction records.

4. The data acquisition method for enhancing large model inference according to claim 1, characterized in that, The construction of the conflict graph based on all the text blocks includes: Based on all the text blocks, obtain the relationship comparison results between each pair of the original texts; the relationship comparison results are contradictory, competing, or parallel. Map each original text corresponding to an information claim to a node in a graph database; Based on the relationship comparison results, generate the relationship edges between each pair of nodes and their strength weights; The relation edges include negative edges, weak negative edges, or positive edges, and the strength weight corresponding to the negative edge is less than the strength weight corresponding to the weak negative edge, and the strength weight corresponding to the weak negative edge is less than the strength weight corresponding to the positive edge.

5. The data acquisition method for enhancing large model inference according to claim 4, characterized in that, The step of determining the conflict confidence level of each node based on the conflict map and all the source confidence levels includes: Configure an initial authority score for each node based on its origin. The relationship edges and their strength weights of the nodes to be analyzed are determined from the conflict graph. Based on the initial authority score and strength weight of the node corresponding to the relation edge of the node to be analyzed, determine the degree of conflict between the node corresponding to the relation edge of the node to be analyzed and the node to be analyzed. Based on all the aforementioned conflict levels and the initial authority score of the node to be analyzed, the conflict credibility of the node to be analyzed is obtained.

6. The data acquisition method for enhancing large model inference according to claim 5, characterized in that, The process of determining the lifecycle state reliability of each node based on the target requirements includes: Determine the lifecycle state corresponding to each node, wherein the lifecycle state includes the existing version time and the corresponding existing version description features; Obtain the current time and current version description features corresponding to the target requirement; Based on the current version description features, the current time, and the lifecycle state, determine the state validity type of each node; Determine the corresponding state weight based on the state validity type; The reliability of the lifecycle state corresponding to the node is determined based on the state weight.

7. The data acquisition method for enhancing large model inference according to claim 5, characterized in that, The step of determining the credibility of the interaction feedback of each node based on the interaction metadata corresponding to each node and the source confidence level includes: Obtain the interaction metadata corresponding to each node; the interaction metadata includes the number of approvals and the number of confirmations. Based on the source confidence level, determine the initial energy information of each node; The credibility of the interaction feedback of each node is determined based on the number of recognitions, the number of confirmations, and the initial energy information.

8. A method for enhancing large-scale RAG inference, characterized in that, include: Retrieve several related original texts from different search sources based on the target requirements; Each of the original text entries is split into several text blocks; Construct a conflict graph based on all the aforementioned text blocks; The conflict graph includes the relationship edges between each pair of the original texts and their strength weights, and each node in the conflict graph is one of the original texts. Determine the source confidence level of the retrieval source for each of the original texts; Based on the conflict map and all the source confidence scores, determine the conflict confidence score for each node; Determine the lifecycle status credibility of each node based on the stated target requirements; The credibility of the interaction feedback of each node is determined based on the interaction metadata corresponding to each node and the source confidence level. Based on the conflict credibility, the lifecycle state credibility, and the interaction feedback credibility, the comprehensive analysis credibility of each node is determined. The original text corresponding to all nodes whose comprehensive analysis credibility is greater than the preset credibility threshold is used as the target data; The target data is then output as the reasoning result through a large model.

9. A computer device, characterized in that, The computer device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the data acquisition method for enhanced large model inference as described in any one of claims 1-7 or the large model inference method for enhanced RAG as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the data acquisition method for enhanced large model inference as described in any one of claims 1-7 or the large model inference method for enhanced RAG as described in claim 8.

11. A computer program product comprising computer instructions, characterized in that, When executed by a processor, the computer instructions implement the data acquisition method for enhanced large model inference as described in any one of claims 1-7 or the large model inference method for enhanced RAG as described in claim 8.

Images