Task information processing method and device, equipment, storage medium and program product

Abstract

The present disclosure provides a task information processing method and device, equipment, storage medium and program product, and relates to the technical field of blockchains. The method comprises: obtaining task request information; using a large language model to analyze the task request information to obtain structured intent data; propagating the signed intent data through a relay node of an intent network; listening to the intent data propagated in the intent network through a solver node of the intent network, solving the intent data to obtain a to-be-executed task corresponding to the task request information; executing the to-be-executed task and returning a task execution result. The method enables a user to propose a natural language task request to perform a blockchain transaction.

Description

Technical Field

[0001] This disclosure relates to the field of blockchain, and more specifically, to a task information processing method, apparatus, electronic device, computer-readable storage medium, and computer program product. Background Technology

[0002] Blockchain systems generally adopt an architecture based on "transactions" as the basic unit, requiring users to explicitly specify counterparties, execution paths, and parameters, resulting in complex interactions and limitations on multi-party collaboration. How to enable users to conduct blockchain transactions using natural language has become a pressing issue.

[0003] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this disclosure is to provide a task information processing method, apparatus, electronic device, readable storage medium, and program product that enables users to conduct blockchain transactions through natural language.

[0005] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part by practice of this disclosure.

[0006] According to one aspect of this disclosure, a task information processing method is provided, comprising: acquiring task request information; parsing the task request information using a large language model to obtain structured intent data; propagating the signed intent data through relay nodes of an intent network; monitoring the intent data propagated in the intent network through solver nodes of the intent network, solving the intent data to obtain a task to be executed corresponding to the task request information; executing the task to be executed, and returning a task execution result.

[0007] According to one embodiment of this disclosure, the structured intent data includes multiple fields, and the data types of the multiple fields include: task type, constraints, budget, expected output, incentive, and time.

[0008] According to one embodiment of this disclosure, a large language model is used to parse the task request information to obtain structured intent data, including: obtaining structured intent prompt information; inputting the task request information and the structured intent prompt information into the large language model to obtain the output result of the large language model; structuring the output result of the large language model into a set of key-value pairs with predetermined text rules; and performing compliance verification on the set of key-value pairs with predetermined text rules to obtain the structured intent data.

[0009] According to one embodiment of this disclosure, the method further includes: visually displaying a summary of the structured intent data to obtain the signed intent data.

[0010] According to one embodiment of this disclosure, propagating the signed intent data through relay nodes of an intent network includes: propagating the signed intent data through relay nodes of the intent network to aggregate the signed intent data into an intent pool; and listening to the intent data propagated in the intent network through solver nodes of the intent network, solving the intent data, and obtaining a task to be executed corresponding to the task request information, including: scanning the intent pool according to the listening frequency through the solver nodes of the intent network, identifying a set of matchable intents using at least one algorithm selected from heuristic search, graph matching algorithm, and optimization planning; and generating a task to be executed that satisfies the constraints of all intent data in the intent set.

[0011] According to one embodiment of this disclosure, the relay nodes of the intent network include a first relay node and a second relay node; the propagation of the signed intent data through the relay nodes of the intent network includes: the first relay node receiving the intent data signed by the user, signing it, and forwarding it to the second relay node; the second relay node receiving the intent data signed by the first relay node, verifying the signature of the first relay node, and signing it again if the verification is successful, and then propagating it.

[0012] According to one embodiment of this disclosure, the method further includes: stimulating the first relay node based on the verification result of the second relay node according to the stimulus in the intent data.

[0013] According to another aspect of this disclosure, a task information processing apparatus is provided, comprising: an acquisition module for acquiring task request information; a parsing module for parsing the task request information using a large language model to obtain structured intent data; a propagation module for propagating the signed intent data through relay nodes of an intent network; a solving module for monitoring the intent data propagated in the intent network through solver nodes of the intent network, solving the intent data, and obtaining a task to be executed corresponding to the task request information; and an execution module for executing the task to be executed and returning a task execution result.

[0014] According to another aspect of this disclosure, an electronic device is provided, comprising: a memory, a processor, and executable instructions stored in the memory and executable in the processor, wherein the processor, when executing the executable instructions, implements any of the methods described above.

[0015] According to another aspect of this disclosure, a computer-readable storage medium is provided that stores computer-executable instructions thereon, which, when executed by a processor, implement any of the methods described above.

[0016] This disclosure provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described above.

[0017] The task information processing method provided in the embodiments of this disclosure obtains task request information, parses the task request information using a large language model to obtain structured intent data, propagates the signed intent data through relay nodes of the intent network, listens to the intent data propagated in the intent network through solver nodes of the intent network, solves the intent data to obtain the task to be executed corresponding to the task request information, executes the task to be executed, and returns the task execution result. Thus, it is possible to enable users to make natural language task requests and conduct blockchain transactions.

[0018] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this disclosure. Attached Figure Description

[0019] The above and other objects, features and advantages of this disclosure will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0020] Figure 1 The illustration illustrates the processing flow of a blockchain network in the relevant technology.

[0021] Figure 2 This is a flowchart illustrating a task information processing method according to an exemplary embodiment.

[0022] Figure 3 It shows Figure 2 The step S204 shown is a schematic diagram of the processing procedure in one embodiment.

[0023] Figure 4 It shows Figure 2 The step S206 shown is a schematic diagram of the processing procedure in one embodiment.

[0024] Figure 5 It shows Figure 2 The step S208 shown is a schematic diagram of the processing procedure in one embodiment.

[0025] Figure 6 according to Figures 2 to 5 An example is shown illustrating the task processing flow that combines a large model with an intent network.

[0026] Figure 7 according to Figures 2 to 6 An example is shown in the diagram of a decentralized collaborative system architecture based on model parsing and intent network propagation.

[0027] Figure 8 A block diagram of a task information processing apparatus according to an embodiment of the present disclosure is shown.

[0028] Figure 9 A block diagram of another task information processing apparatus according to an embodiment of the present disclosure is shown.

[0029] Figure 10 A schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0030] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.

[0031] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, apparatuses, steps, etc., can be employed. In other instances, well-known structures, methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0032] Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this disclosure, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. The symbol " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0033] In this disclosure, unless otherwise expressly specified and limited, the term "connection" and similar terms should be interpreted broadly, for example, it can refer to an electrical connection or the ability to communicate with each other; it can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this disclosure according to the specific circumstances.

[0034] The following is an explanation of the terms used in this disclosure.

[0035] Large Language Model (LLM): An artificial intelligence model with natural language understanding and generation capabilities (such as GPT, Claude, Deepseek, etc.).

[0036] Blockchain: A distributed ledger technology that uses cryptographic principles to build and maintain an immutable, decentralized data structure.

[0037] Intent: A user's expressed goal or task request within the system. It can be incomplete or open, requiring collaboration from other nodes in the network to complete. Intents are represented in structured data form and carry information such as constraints, preferences, and incentives.

[0038] Intent Gossip Layer: A decentralized network protocol designed for intent propagation, discovery, and matching, allowing nodes to store, index, combine, and solve intents based on content. Intents can be broadcast across the network via multi-hops and support path authentication and incentives.

[0039] Structured Intent: An intent object generated based on a predefined JSON Schema. It can contain fields such as task type, resource constraints, budget, incentives, and time. It can be parsed, signed, matched, and executed by various modules within the system.

[0040] As mentioned above, since blockchain network nodes are only responsible for verification and execution, and thousands of nodes must reach a completely consistent consensus on the transaction results, the input and logic of blockchain network transactions must be completely self-contained and unambiguous. Figure 1 The illustration schematically depicts the processing flow of a blockchain network in related technologies. For example... Figure 1 As shown, transaction 104 constructed by user 102 enters the mempool (S106), meaning that this transaction has been received by a node in the blockchain network. An example of transaction 104 constructed by user 102 is as follows: "nonce": 7, "to": "0x4bbe ... 261520", "value": "0x2386f26fc10000", "gasPrice": "0x09184e72a000", "gasLimit": "0x27100", "data": "Ox", "chainId": 1, "v": "...", "r": "...", "s": "..." In this context, the "to" field represents the target address of the transaction, the "value" field represents the amount of ETH to be sent, the "gasPrice" field represents the price the user is willing to pay per unit of Gas, the "gasLimit" field represents the maximum amount of Gas the user is willing to pay for this transaction, the "data" field represents the call information passed to the smart contract, the "chainId" field represents the identifier of the blockchain network, and the "v", "r", and "s" fields represent the cryptographic signature generated by the sender's private key.

[0041] After the transactions in the mempool are verified, consensus (PoW or PoS) determines which node's submitted transaction can become the recognized truth (i.e., S108. Consensus). Then, the selected transaction is written into a block (S110), and then verified and confirmed by all nodes in the network, and the transaction is completed (S112).

[0042] It is evident that if users directly utilize blockchain networks for transactions, their transaction requests must explicitly specify the receiving address or contract, each operational step, all parameters, and so on. This demands a high level of technical knowledge from users and is cumbersome and error-prone. While large-scale models possess powerful natural language parsing and abstraction capabilities, they are difficult to embed into decentralized semantic protocols. Intent networks, as a novel system architecture, introduce the concept of "intent," allowing users to declare only their target state or preferences, with the network automatically handling counterparty discovery, matching, and transaction execution. However, achieving seamless integration between natural language and intent networks remains a critical challenge.

[0043] Therefore, this disclosure proposes a natural language task-driven system architecture that combines a large language model with an intent network, enabling users to express intentions such as transactions, exchanges, executions, and service requests through natural language. The system then parses these intentions into structured intentions and completes decentralized matching, collaboration, and execution through an intent propagation network.

[0044] Figure 2This is a flowchart illustrating a task information processing method according to an exemplary embodiment. For example... Figure 2 The method shown can be applied, for example, to intent networks.

[0045] refer to Figure 2 The method 20 provided in this embodiment may include the following steps.

[0046] In step S202, task request information is obtained.

[0047] In some embodiments, the task request information can be a request message provided by the user in natural language form, specifying the task objective. For example, user A sends a task request message: "I want to sell my NFT #12345 and hope to exchange it for 100-120 USDC." Another example is user B sending a task request message: "I have 100 USDC and want to buy an NFT (without specifying an ID)." Yet another example is user C sending a task request message: "I am willing to pay 110 USDC to buy NFT #12345."

[0048] In step S204, the task request information is parsed using a large language model to obtain structured intent data.

[0049] In some embodiments, the LLM can parse the task request information sent by the user in natural language into structured intent data, which may be, for example, an intent object. The structured intent data may include multiple fields, and the data types of these fields may include: task type, constraints, budget, expected output, incentive, and time.

[0050] In some embodiments, prompts suitable for structured intent data can be constructed and input into the LLM along with task request information. The output of the LLM is then processed to obtain structured intent data. Exemplary implementations can be referred to... Figure 3 .

[0051] In some embodiments, the construction intent is a first type of interaction object, and the intent may be a partial, incomplete, or supplementary declaration.

[0052] In step S206, the signed intent data is propagated through the relay nodes of the intent network.

[0053] In some embodiments, the intent network can be a sparse virtual overlay network built on libp2p, where all nodes support message forwarding (relay nodes), and some nodes act as solvers (solver nodes) and intent caches. The network supports content addressing and filtering based on semantic dimensions such as task type and resource requirements. The intent network supports decentralized, multi-hop, and latency-tolerant intent broadcasting, ensuring broad coverage and high availability of intents within the network. The intent network can also support cost sharing to incentivize nodes to participate in forwarding (e.g., each node on the path receives a share of the propagation incentive). The intent network is the data propagation and matching infrastructure of this system, designed to support decentralized, multi-target, and latency-tolerant intent exchange and matching.

[0054] In some embodiments, each signed intent data may include a unique ID, author signature, intent data structure, expiration time, and incentive mechanism. Intent broadcasting can employ a path signature mechanism, where nodes can prove the message chain, preventing message tampering and forgery. Exemplary implementations can be found in [reference needed]. Figure 4 .

[0055] In some embodiments, signed intent data is propagated through relay nodes in the intent network to aggregate the signed intent data into an intent pool. After being broadcast to the network by the propagation nodes, the intent is aggregated into the intent pool for solvers to acquire and compete for.

[0056] In step S208, the intent data propagated in the intent network is monitored by the solver node of the intent network, the intent data is solved, and the task to be executed corresponding to the task request information is obtained.

[0057] In some embodiments, heuristic search, graph matching, optimization planning, and other algorithms can be combined to support bilateral (e.g., token exchange), trilateral (e.g., barter), and multilateral matching (e.g., group auctions). Exemplary implementations can be found in [reference needed]. Figure 5 .

[0058] In some embodiments, invisible preferences can be declared in the intent, and solver nodes can provide zero-knowledge proofs to ensure matching fairness and privacy.

[0059] In step S210, the task to be executed is performed, and the task execution result is returned.

[0060] In some embodiments, encrypted transactions (tasks to be executed) are sorted by the blockchain's consensus system, decrypted at a threshold, and processed by the execution environment. The processing result (task execution result) can be sent back to the user via state updates or event channels. The blockchain can be used to store evaluation data hashes and record evaluation results, ensuring the transparency and immutability of the evaluation process.

[0061] According to the task information processing method provided in this disclosure, task request information is obtained, and the task request information is parsed using a large language model to obtain structured intent data. The signed intent data is propagated through the relay nodes of the intent network, and the intent data propagated in the intent network is monitored by the solver nodes of the intent network. The intent data is solved to obtain the task to be executed corresponding to the task request information. Then, the task to be executed is executed, and the task execution result is returned. Thus, it is possible to realize blockchain transactions as long as the user makes a natural language task request, realizing an integrated task lifecycle closed loop from natural language to on-chain executable state, and ensuring the structural security, semantic consistency and execution feasibility of the generated result.

[0062] Figure 3 It shows Figure 2 The diagram shown illustrates step S204 in one embodiment. Figure 3 As shown in this embodiment of the disclosure, step S204 may further include steps S302 to S308. Step S310 may be performed after step S308.

[0063] Step S302: Obtain structured intent prompt information.

[0064] In some embodiments, prompts can be constructed based on task templates, thereby making the format of the content generated by the LLM correspond to structured intent data. The structured intent prompts can define which fields the final output should include, the types of these fields, and their relationships, such as the task type, resource constraints, budget, and other fields mentioned above.

[0065] Step S304: Input the task request information and structured intent prompt information into the large language model to obtain the output of the large language model.

[0066] In some embodiments, the instructions given to the LLM may include, in addition to describing the task by entering task request information, several completed samples, i.e., the prompt information mentioned above, to demonstrate the desired output format and quality.

[0067] Step S306: The output of the large language model is structured into a set of key-value pairs with predetermined text rules.

[0068] In some embodiments, components such as a JSON format extractor can be used to structure the LLM output into a JSON object. An example of structured intent data is shown below: { "intent_id": "a5f2c1e3-78bc-4e5f-b2f1-1c03a6fd0d92", "author": "0xAbC123...789DeF", "task_type": "nft_swap", "asset_out": { "type": "ERC-721", "token_id": "12345" }, "asset_in": { "type": "ERC-20", "range": "100-120 USDC" }, "constraints": { "ratio_tolerance": "±5%" }, "incentive": { "type": "payment", "amount": "110 USDC" }, "deadline": "2025-08-10T00:00:00Z", "signature": null } Step S308: Perform compliance verification on the set of key-value pairs with predetermined text rules to obtain structured intent data.

[0069] In some embodiments, components such as schema validators can be used to validate JSON objects to ensure the legality of fields and formats; Step S310: Visualize the summary of the structured intent data to obtain the signed intent data.

[0070] In some embodiments, a summary of the generated structured intent data can be visualized for user confirmation of the signature. The intent data generated by LLM can be signed by a local agent or wallet plugin.

[0071] According to the method provided in the embodiments of this disclosure, a large language model is used to automatically parse the natural language request input by the user into a structured intent. The generated intent conforms to a verifiable JSON Schema and can be used for on-chain task expression. It also supports task type recognition, parameter extraction, constraint modeling and preference modeling, which significantly reduces the threshold for user interaction: users do not need to understand complex transaction formats and can initiate on-chain tasks only through natural language, thus reducing the threshold for Web3 applications.

[0072] Figure 4 It shows Figure 2 The step S206 shown is a schematic diagram of the processing procedure in one embodiment. (See attached diagram.) Figure 4 As shown in this embodiment, step S206 may further include steps S402 to S404, wherein the first relay node and the second relay node may both be... Figure 2 The relay node of the target network. Step S406 is related to step S404 and can be performed after the task is executed.

[0073] In step S402, the first relay node receives the intent data after the user signs it, signs it, and forwards it to the second relay node.

[0074] In step S404, the second relay node receives the intent data signed by the first relay node, verifies the signature of the first relay node, and, if the verification is successful, signs the data and then propagates it.

[0075] In some embodiments, each relay node signs its transmission path when forwarding, forming a transmission chain. Any node that receives the intent can verify this chain to confirm that the intent was indeed sent from an original node and has not been modified or forged along the way.

[0076] For example, suppose user A sends an intent, which is forwarded to node D through nodes B and C: node B will sign the intent after receiving it before forwarding it; node C will also sign the intent after receiving it from B before forwarding it; when node D receives the intent, it can verify the complete signature chain A → B → C → D.

[0077] Step S406: Incentivize the first relay node based on the verification result of the second relay node according to the incentive in the intent data.

[0078] In some embodiments, if a node maliciously tampers with the content when forwarding an intent, its signature will not pass the verification of downstream nodes, causing subsequent nodes to refuse to propagate the intent, and the node will also lose its incentive eligibility.

[0079] According to the method provided in the embodiments of this disclosure, by designing an intent propagation protocol with path authentication, secure relay and node incentive mechanisms, the relay node not only participates in forwarding, but also has identity authentication capabilities, and ensures that its behavior is honest through the incentive mechanism, thus ensuring the integrity and traceability of each intent propagated in the P2P network, forming an incentive-compatible secure relay mechanism.

[0080] Figure 5 It shows Figure 2 The step S208 shown is a schematic diagram of the processing procedure in one embodiment. (See attached diagram.) Figure 5 As shown in the present embodiment, step S208 may further include steps S502 to S504.

[0081] Step S502: The intent pool is scanned by the solver nodes of the intent network according to the listening frequency, and at least one of the following algorithms is used to identify the set of matching intents: heuristic search, graph matching algorithm, and optimization planning.

[0082] Step S504: Generate a task to be executed that satisfies the constraints of all intent data in the intent set.

[0083] In some embodiments, the system does not require all intentions to be fully explicit, but rather allows them to "openly declare partial goals," which are then matched using algorithms such as heuristic search, graph matching, and optimization planning. For example, a user might input: "Just help me exchange what I have for usable tokens." Based on this, the LLM generates: { "task_type": "token_swap", "asset_out": {"type": "ERC-20", "amount": "200"}, "asset_in": {"type": "ERC-20", "constraint": "can be used to pay gas fees"}, ... } Although the input is highly ambiguous, LLM can extract task intent and some structured elements, providing a starting point for the matching algorithm. The system then uses a graph structure to combine these "incomplete intents" into a transaction or task flow that can be executed in a closed loop. This is a key advantage of intent networks compared to peer-to-peer matching systems in related technologies.

[0084] For example: User A expresses the intention: I want to sell one of my NFTs and hope to get 100-120 USDC in return; User B expresses the intention: I have 100 USDC and want to buy an NFT, but I haven't specified the serial number; User C expresses the intention: I'm willing to pay 110 USDC to buy NFT #12345. The system can recognize the compatibility between these intentions and can automatically combine them into a closed-loop transaction even if the user does not explicitly provide the target transaction object.

[0085] According to the method provided in the embodiments of this disclosure, the intent network supports incomplete intent matching and combination solving, which solves the problem that on-chain tasks are difficult to self-organize and coordinate, and can improve multi-party collaboration capabilities and enhance resource matching efficiency.

[0086] Figure 6 according to Figures 2 to 5 This example illustrates a task processing flow that combines a large model with an intent network. For example... Figure 6 As shown, the large language model automatically parses the natural language request input by user 602 into a structured intent with semantic integrity and machine verifiability. The intent is then broadcast, stored, discovered, and combined through an intent propagation network (S604) to form an executable state transition. Solver nodes then automatically listen for and execute the combination and solution (S606), constructing a transaction that conforms to system constraints and entering the memory pool (S608). After the transactions in the memory pool are verified, consensus (PoW or PoS) determines which node's submitted transaction can become the accepted truth (S610, consensus). The selected transaction is then written into a block (S612), and verified and confirmed by all nodes in the network, completing the transaction (S614).

[0087] Figure 7 according to Figures 2 to 6 An example is shown in the diagram of a decentralized collaborative system architecture based on model parsing and intent network propagation. For example... Figure 7 As shown, the transaction process based on this system architecture may include the following steps S7002 to S7006.

[0088] Step S7002, Counterparty Discovery. In this stage, user 702 creates an intent using natural language and inputs it into the LLM. The LLM generates a structured intent including state constraints, matching preferences, budget, etc. The intent is broadcast through the P2P network 706 and is cached and indexed. Solver 708 listens to the intent pool, runs a solving algorithm to identify composable intent sets, generates declarative transactions that satisfy all intent constraints, and aggregates them into the transaction mempool 710.

[0089] Intents can declare invisible preferences, with zero-knowledge proofs ensuring matching fairness and privacy. For example, they can be transparent intents (tl), where all parameters (asset, quantity, price) are publicly broadcast in plaintext ([tl1, tl2, tl3, tl4]). They can be semi-privacy intents (sl), where users can choose to hide some sensitive fields in the intent. Matching fairness can be guaranteed through basic cryptographic commitments and zero-knowledge proofs (ZK proofs), ensuring that the solver can verify that the bid satisfies the user's hidden constraints even without knowing the specific values ​​of ([sl1, sl2, sl3, sl4]). They can be private intents (pl), where users can hide not only specific parameters but also the type of intent and asset class itself. Their core implementation relies on zero-knowledge proof technology. The solver or matching engine can verify the matching feasibility between multiple private intents ([pl1, pl2, pl3, pl4]) through recursive ZK proofs without knowing the user's privacy, thus achieving fair and effective global matching while ensuring absolute privacy.

[0090] Step S7004, Consensus. In this stage, proposer 712 is selected by a consensus algorithm (such as random selection in PoS or computational competition in PoW) and filters transactions to be processed from the transaction mempool 710 (e.g., Tx corresponding to the aforementioned intentions). c Tx b Tx a The process involves verifying and packaging the block, then constructing a block, signing it, and broadcasting it. Validator 714, representing the majority of nodes in the network, independently reviews the block submitted by proposer 712 and collectively votes to decide whether to accept it, thus obtaining the final block. This phase verifies data availability and security and determines the concurrency domain.

[0091] Step S7006, Execution and Verification. In this stage, the entire process runs in a unified execution environment capable of handling states corresponding to three different privacy levels of intent: transparent, semi-privacy, and private. Regardless of the privacy level of the intent, the resulting "state transitions" (i.e., changes in on-chain states such as user balances and contract storage) must undergo rigorous validity verification. This verifies that the transition from the old state to the new state fully complies with all constraints of the intent declaration (e.g., the user has indeed received no less than the promised amount of tokens).

[0092] Once a state transition is verified as valid, its digest (usually represented as a new state root) enters the public verification phase. This means that any network node can independently verify the correctness of the state transition. Subsequently, this verified state root is included as a credential in the next blockchain block. When the block containing this state root achieves network consensus and is finally confirmed, the corresponding state transition obtains its settled final state. At this point, the "intended settlement" is completed and publicly verified by the entire network, and its result becomes part of the immutable blockchain history.

[0093] Cryptographic transactions are sorted by the consensus system. After threshold decryption, they are processed by the execution environment. The results are then sent back to the user (718) via state updates or event channels. The user receives the matching feedback and results and confirms the transaction accordingly.

[0094] According to the method provided in this disclosure, a structured intent format supporting features such as field integrity verification, author signature, incentive binding, and timeout expiration is constructed by designing a structured intent construction and signature mechanism. This allows for matching, verification, and combination without exposing the original semantics. The structured intent possesses strong constraints and verifiability, enhancing protocol expressiveness. It also supports zero-knowledge hiding, budget binding, and incentive declarations, improving interaction security and incentive design space. The system supports embedding into multiple chains, models, and interaction methods (CLI, wallet, natural language API), facilitating ecosystem integration and standard evolution. It exhibits strong adaptability and good standardization. This system achieves native embedding of language models at the blockchain protocol layer, making AI capabilities part of the protocol logic, pioneering a new paradigm for intelligent service networks, and promoting the native integration of LLM and blockchain.

[0095] Figure 8 This is a block diagram illustrating a task information processing apparatus according to an exemplary embodiment. Figure 8 The task information processing device shown can be applied, for example, to an intent network.

[0096] refer to Figure 8 The apparatus 80 provided in this embodiment may include an acquisition module 802, a parsing module 804, a propagation module 806, a solving module 808, and an execution module 810.

[0097] The acquisition module 802 can be used to acquire task request information.

[0098] The parsing module 804 can be used to parse task request information using a large language model to obtain structured intent data.

[0099] The propagation module 806 can be used to propagate signed intent data through relay nodes of the intent network.

[0100] The solver module 808 can be used to listen to the intent data propagated in the intent network through the solver node of the intent network, solve the intent data, and obtain the task to be executed corresponding to the task request information.

[0101] The execution module 810 can be used to execute tasks to be executed and return the task execution results.

[0102] Figure 9 This is a block diagram illustrating another task information processing apparatus according to an exemplary embodiment. Figure 9 The task information processing device shown can be applied, for example, to an intent network.

[0103] refer to Figure 9 The apparatus 90 provided in this embodiment may include an acquisition module 902, a parsing module 904, a display module 905, a propagation module 906, a solving module 908, an execution module 910, and a settlement module 912.

[0104] The acquisition module 902 can be used to acquire task request information.

[0105] The parsing module 904 can be used to parse task request information using a large language model to obtain structured intent data.

[0106] Structured intent data can include multiple fields, and the data types of these fields include: task type, constraints, budget, expected output, incentive, and time.

[0107] The parsing module 904 can also be used to: obtain structured intent prompt information; input task request information and structured intent prompt information into the large language model to obtain the output result of the large language model; structure the output result of the large language model into a set of key-value pairs with predetermined text rules; and perform compliance verification on the set of key-value pairs with predetermined text rules to obtain structured intent data.

[0108] Display module 905 can be used to visualize a summary of structured intent data to obtain signed intent data.

[0109] The propagation module 906 can be used to propagate signed intent data through relay nodes in the intent network.

[0110] The relay nodes of the intent network can include a first relay node and a second relay node.

[0111] The propagation module 906 can also be used to propagate signed intent data through relay nodes of the intent network to aggregate the signed intent data into the intent pool.

[0112] The propagation module 906 can also be used for: the first relay node receiving the intent data signed by the user, signing it, and forwarding it to the second relay node; the second relay node receiving the intent data signed by the first relay node, verifying the signature of the first relay node, and signing it and propagating it if the verification is successful.

[0113] The solver module 908 can be used to listen to the intent data propagated in the intent network through the solver node of the intent network, solve the intent data, and obtain the task to be executed corresponding to the task request information.

[0114] The solver module 908 can also be used to: scan the intent pool according to the listening frequency through the solver nodes of the intent network, identify the set of matching intents using at least one of heuristic search, graph matching algorithm, and optimization planning; and generate tasks to be executed that satisfy the constraints of all intent data in the intent set.

[0115] The execution module 910 can be used to execute tasks and return the task execution results.

[0116] The settlement module 912 can be used to incentivize the first relay node based on the verification result of the second relay node according to the incentive in the intent data.

[0117] The specific implementation of each module in the apparatus provided in this embodiment can be referred to the content of the above method, and will not be repeated here.

[0118] Figure 10 A schematic diagram of the structure of an electronic device according to an embodiment of this disclosure is shown. It should be noted that... Figure 10 The devices shown are merely examples of computer systems and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0119] like Figure 10 As shown, device 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1002 or a program loaded from storage section 1008 into random access memory (RAM) 1003. RAM 1003 also stores various programs and data required for the operation of device 1000. CPU 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Input / output (I / O) interface 1005 is also connected to bus 1004.

[0120] The following components are connected to I / O interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to I / O interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 1010 as needed so that computer programs read from it can be installed into storage section 1008 as needed.

[0121] 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 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 communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by central processing unit (CPU) 1001, it performs the functions defined above in the system of this disclosure.

[0122] It should be noted that the computer-readable medium disclosed herein may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0123] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0124] The modules described in the embodiments of this disclosure can be implemented in software or hardware. The described modules can also be housed in a processor; for example, a processor can be described as including an acquisition module, a parsing module, a propagation module, a solving module, and an execution module. The names of these modules do not necessarily limit the module itself; for example, the acquisition module can also be described as "a module for acquiring task request information provided by a user in natural language."

[0125] In another aspect, this disclosure also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to include: Obtain task request information; parse the task request information using a large language model to obtain structured intent data; propagate the signed intent data through relay nodes in the intent network; listen to the intent data propagated in the intent network through solver nodes in the intent network, solve the intent data, and obtain the task to be executed corresponding to the task request information; execute the task to be executed and return the task execution result.

[0126] This disclosure provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described above.

[0127] Exemplary embodiments of this disclosure have been specifically shown and described above. It should be understood that this disclosure is not limited to the detailed structures, arrangements, or implementations described herein; rather, this disclosure is intended to cover various modifications and equivalent arrangements contained within the spirit and scope of the appended claims.

Claims

1. A task information processing method, characterized in that, include: Obtain task request information; The task request information is parsed using a large language model to obtain structured intent data; The signed intent data is propagated through relay nodes in the intent network; The solver node of the intent network listens to the intent data propagated in the intent network, solves the intent data, and obtains the task to be executed corresponding to the task request information. Execute the task to be executed and return the task execution result.

2. The method according to claim 1, characterized in that, The structured intent data includes multiple fields, and the data types of these fields include: task type, constraints, budget, expected output, incentive, and time.

3. The method according to claim 1 or 2, characterized in that, The task request information is parsed using a large language model to obtain structured intent data, including: Obtain structured intent prompts; The task request information and the structured intent prompt information are input into the large language model to obtain the output of the large language model; The output of the large language model is structured into a set of key-value pairs with predetermined text rules; The set of key-value pairs with predetermined text rules is subjected to compliance verification to obtain the structured intent data.

4. The method according to claim 1 or 2, characterized in that, Also includes: A summary of the structured intent data is visualized to obtain the signed intent data.

5. The method according to claim 1 or 2, characterized in that, The signed intent data is propagated through relay nodes in the intent network, including: The signed intent data is propagated through relay nodes in the intent network to aggregate the signed intent data into the intent pool. The solver node of the intent network listens to the intent data propagated in the intent network, solves the intent data, and obtains the tasks to be executed corresponding to the task request information, including: The solver nodes of the intent network scan the intent pool according to the listening frequency, and identify the set of matching intents using at least one of the following algorithms: heuristic search, graph matching algorithm, and optimization planning. Generate an executable task that satisfies the constraints of all intent data in the intent set.

6. The method according to claim 1 or 2, characterized in that, The relay nodes of the intent network include a first relay node and a second relay node; The signed intent data is propagated through relay nodes in the intent network, including: The first relay node receives the intent data after the user has signed it, signs it, and then forwards it to the second relay node; The second relay node receives the intent data signed by the first relay node, verifies the signature of the first relay node, and, if the verification is successful, signs the first relay node and then propagates it.

7. The method according to claim 6, characterized in that, Also includes: The first relay node is incentivized based on the verification result of the second relay node according to the incentive in the intent data.

8. A task information processing device, characterized in that, include: The acquisition module is used to obtain task request information; The parsing module is used to parse the task request information using a large language model to obtain structured intent data; The propagation module is used to propagate the signed intent data through relay nodes of the intent network; The solving module is used to listen to the intent data propagated in the intent network through the solver node of the intent network, solve the intent data, and obtain the task to be executed corresponding to the task request information. The execution module is used to execute the task to be executed and return the task execution result.

9. An electronic device, comprising: A memory, a processor, and executable instructions stored in the memory and executable in the processor, characterized in that the processor, when executing the executable instructions, implements the method as described in any one of claims 1-7.

10. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that, When the executable instructions are executed by the processor, they implement the method as described in any one of claims 1-7.

11. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method as described in any one of claims 1-7.

Images