Ang isang patuloy na hamon sa malalaking modelo ng wika (LLMs) ay ang kanilang tendensya na makagawa ng output na hindi mahuhulaan. Sa kabila ng maingat na ginawang mga prompt, ang mga LLM ay madalas na lumilihis sa mga inaasahan, na ginagawang mahirap ang kanilang output na muling gamitin o isama sa mga daloy ng trabaho. Ang mas masahol pa, ang mga resulta ay madalas na hindi nauulit, na nagpapalubha sa downstream na pagproseso. Sa post sa blog na ito, tinutuklasan namin , isang diskarteng dynamic na bumubuo ng mga JSON schema para sa mga solusyon bago magsagawa ng mga gawain. Nakakatulong ito na lumikha ng mas magagamit muli, maaasahan, at mahuhulaan na mga prompt. ang structured meta-prompting Background Karamihan sa mga modernong LLM ay nag-aalok na ngayon ng JSON output mode, ngunit ang mga resulta ay hindi palaging umaayon sa isang inaasahang schema. Kadalasang ginagamit ng mga developer ang mga mekanismo ng check-and-retry, na nakakaubos ng oras, mahal, at madaling masira ang karanasan ng user. Sa , ipinakilala ng OpenAI ang mga structured na output na ginagarantiyahan na ang output ay aayon sa isang JSON schema na ibinigay ng user. Bagama't hindi isiniwalat ang mga eksaktong detalye, malamang na nakakamit ito ng kumbinasyon ng limitadong pag-decode at pre-sample na logit biasing. Ang mga katulad na kakayahan ay inaalok ng iba pang mga tool tulad ng patnubay, mga balangkas, pagbabaligtad, CommandR, SGLang. Ang mga diskarteng ito ay nagbibigay ng kapangyarihan sa mga developer na gabayan ang mga output sa mga JSON schema, bawat isa ay may natatanging trade-off. gpt4o Ano ang structured meta-prompting? Ang structured meta-prompting ay anumang pamamaraan na gumagamit ng LLM para dynamic na gumawa ng structured na paglalarawan ng isang problema bago gumawa ng solusyon. Ang pamamaraang ito ay nag-aalok ng ilang mga pakinabang kaysa sa direktang pagdikta: : Binubuo ang mga istruktura ng output sa runtime batay sa mga paglalarawan ng gawain, sa halip na maging hardcoded o fine-tune sa modelo. Dynamic : Ang nabuong schema ay nababasa ng tao at maaaring siyasatin, i-verify, o baguhin ng mga tao o iba pang LLM. Adaptable : Maaaring i-save at muling gamitin ang schema sa maraming gawain, pagpapatakbo, at makina. Magagamit muli : Ang mga solusyon ay aayon sa isang mahusay na tinukoy na istraktura, na ginagawang angkop ang mga ito para sa downstream na pagkalkula. Mahuhulaan Code Maglakad tayo sa isang halimbawa kung saan gumagamit tayo ng structured meta-prompting para gumawa ng outline para sa isang bagong bestselling spy thriller novel. Ang lahat ng code para sa blog post na ito ay available sa na ito. Colab Notebook tiyaking itakda ang iyong sa Colab Secrets OPENAI_API_KEY 1. Pagbuo ng isang istraktura dAng unang hakbang ay upang tukuyin ang isang JSON schema na naglalarawan sa nais na output. Upang gawin ito, gagawa kami ng JSON schema para sa isang JSON schema — epektibo, isang ! Ibinigay ito bilang bahagi ng detalye ngunit kailangan nating gumawa ng ilang pagbabago. meta-schema 1.1 Pag-aayos ng meta-schema Ang mga structured output API ng OpenAI at Cohere ay nagpapataw sa mga karaniwang schema ng JSON. Ito ay medyo nakakalungkot ngunit isang bagay na maaari nating ayusin sa ngayon. I-tweak namin ang meta-schema para matiyak ang pagiging tugma: ng ilang mga paghihigpit from jsonschema import Draft202012Validator def openai_compatible_metaschema(schema: Dict[str, object]): schema["type"] = "object" del schema["allOf"] return schema openai_json_metaschema = openai_compatible_metaschema( copy.deepcopy(Draft202012Validator.META_SCHEMA) ) *: Tila* at sumusuporta sa mga arbitrary na JSON schema...ngunit wala pang pampublikong access! Tandaan walang ganoong mga paghihigpit ang inversion 1.2 Pagdaragdag ng mga paghihigpit sa aming meta-prompt Susunod, isinasama namin ang mga alituntunin sa prompt upang matiyak na ang resultang JSON schema ay sumusunod sa mga hadlang ng OpenAI. Halimbawa: Kinakailangan ang lahat ng field. Ang mga bagay ay may mga limitasyon sa lalim at laki ng pugad. Ang mga karagdagang katangian ay dapat na hindi pinapayagan ( ). "additionalProperties": false system_guidelines = "\n".join( [ "All fields must be required - To use Structured Outputs, all fields or function parameters must be specified as required. NOTE: Although all fields must be required (and the model will return a value for each parameter), it is possible to emulate an optional parameter by using a union type with null." "Objects have limitations on nesting depth and size - A schema may have up to 100 object properties total, with up to 5 levels of nesting.", "Limitations on total string size - In a schema, total string length of all property names, definition names, enum values, and const values cannot exceed 15,000 characters.", "Limitations on enum size - A schema may have up to 500 enum values across all enum properties. For a single enum property with string values, the total string length of all enum values cannot exceed 7,500 characters when there are more than 250 enum values.", "additionalProperties: false must always be set in objects - additionalProperties controls whether it is allowable for an object to contain additional keys / values that were not defined in the JSON Schema. Structured Outputs only supports generating specified keys / values, so we require developers to set additionalProperties: false to opt into Structured Outputs.", "Some type-specific keywords are not yet supported - Notable keywords not supported include: For strings: minLength, maxLength, pattern, format; For numbers: minimum, maximum, multipleOf; For objects: patternProperties, unevaluatedProperties, propertyNames, minProperties, maxProperties; For arrays: unevaluatedItems, contains, minContains, maxContains, minItems, maxItems, uniqueItems", "For anyOf, the nested schemas must each be a valid JSON Schema per this subset", "Definitions are supported - You can use definitions to define subschemas which are referenced throughout your schema. The following is a simple example.", "Recursive schemas are supported - Sample recursive schema using # to indicate root recursion.", ] ) 1.3 Pagse-set up ng meta-prompt Tinutukoy na namin ngayon ang meta-prompt na bumuo ng JSON schema para sa spy thriller outline from langchain.prompts import ChatPromptTemplate task_description = "Write an outline for a bestselling spy thriller novel" task_guidelines = """ - You must follow one of the six basic story arcs: Rags to riches, Riches to rags, Icarus, Oedipus, Cinderella, Man in a hole - Outputs must include characters, plot points (including exposition, rising action, climax, falling action, and resolution), central conflict, setting, major turning points or "beats," character arcs, and a synopsis of the story; essentially, a detailed breakdown of the key elements that will drive the narrative throughout the novel. """ prompt_messages = ChatPromptTemplate.from_messages( [ ( "system", "You are an expert in creating JSON schemas. You have been asked to generate a detailed JSON schema for the output of a given task based based on a task desciption and some guidelines.", ), ( "system", "Your JSON schema must always adhere to the following system system guidelines for JSON schemas:\n<system_guidelines>\n{system_guidelines}\n</system_guidelines>", ), ( "user", "Use the task description and guidelines below to generate an output JSON schema for the following task based on the guidelines provided.\n\n<task_description>\n{task_description}\n</task_description>\n\n<guidelines>\n{task_guidelines}\n</guidelines>", ), ] ) 1.4 Pagbuo ng istraktura Hinihimok namin ang LLM upang lumikha ng schema: messages = prompt_messages.format_messages( system_guidelines=system_guidelines, task_description=task_description, task_guidelines=task_guidelines ) ## Make sure to set up OPENAI_API_KEY in your Colab Secrets ## https://x.com/GoogleColab/status/1719798406195867814 client = OpenAI(api_key=userdata.get('OPENAI_API_KEY')) model = "gpt-4o" metaprompt_completion = client.beta.chat.completions.parse( model=model, messages=convert_to_penai_messages(messages), response_format={ "type": "json_schema", "json_schema": JSONSchema( name="JsonMetaschema", description="JSON Metaschema for the 2020-12 Draft of the JSON Schema specification that can be used to validate JSON data", schema=openai_json_metaschema, strict=False, ) } ) task_output_schema = json.loads(metaprompt_completion.choices[0].message.content) print(json.dumps(task_output_schema, indent=2)) { "$schema": "https://json-schema.org/draft/2020-12/schema", "title": "Outline for a Bestselling Spy Thriller Novel", "type": "object", "properties": { "storyArc": { "type": "string", "enum": [ "Rags to riches", "Riches to rags", "Icarus", "Oedipus", "Cinderella", "Man in a hole" ] }, "characters": { "type": "array", "items": { "type": "object", "properties": { "name": { "type": "string" }, "role": { "type": "string" }, "description": { "type": "string" }, "arc": { "type": "string" } }, "required": ["name", "role", "description", "arc"], "additionalProperties": false }, "minItems": 1 }, "plotPoints": { "type": "object", "properties": { "exposition": { "type": "string" }, "risingAction": { "type": "string" }, "climax": { "type": "string" }, "fallingAction": { "type": "string" }, "resolution": { "type": "string" } }, "required": [ "exposition", "risingAction", "climax", "fallingAction", "resolution" ], "additionalProperties": false }, "centralConflict": { "type": "string" }, "setting": { "type": "string" }, "majorTurningPoints": { "type": "array", "items": { "type": "string" }, "minItems": 1 }, "characterArcs": { "type": "object", "properties": { "protagonistArc": { "type": "string" }, "antagonistArc": { "type": "string" }, "supportingCharactersArcs": { "type": "array", "items": { "type": "string" }, "minItems": 0 } }, "required": [ "protagonistArc", "antagonistArc", "supportingCharactersArcs" ], "additionalProperties": false }, "synopsis": { "type": "string" } }, "required": [ "storyArc", "characters", "plotPoints", "centralConflict", "setting", "majorTurningPoints", "characterArcs", "synopsis" ], "additionalProperties": false } Mayroon na kaming schema na naglalarawan sa outline para sa aming spy thriller novel. Ito ay maaaring ipagpatuloy sa isang file o sa isang database. 2. Pagbuo ng solusyon 2.1 Mabilis na pag-setup Gamit ang schema, tinukoy namin ang isang prompt para sa balangkas ng nobela. gagamit kami ng ilang pangunahing pag-udyok at mga alituntunin sa role-play: user_requirements = "Tell a story about counter-intelligence operative working against the clock. The novel should be extremely realistic, slow burn." task_prompt = ChatPromptTemplate.from_messages( [ ( "system", "You are a world-renowned author that has written dozens of bestselling thriller novels. Your task is to create an outline for a new novel based on the user's requirements.", ), ( "user", "Please write a novel outline based strictly on the following requirements <requirements>{requirements}</requirements>", ), ] ) task_completion = client.beta.chat.completions.parse( model=model, messages=convert_to_openai_messages(task_prompt.format_messages(requirements=user_requirements)), response_format={ "type": "json_schema", "json_schema": JSONSchema( ## TODO: You can change this depending your task name="ThrillerNovelOutlineSchema", description="A schema for outlining a new novel", schema=task_output_schema, strict=False, ) } ) task_result = json.loads(task_completion.choices[0].message.content) Narito ang outline para sa aming susunod na spy thriller: { "storyArc": "Cinderella", "synopsis": "In 'The Clockwork Veil', Ethan Cross, a savvy counter-intelligence operative, is thrust into a high-pressure scenario where leaked documents threaten national integrity. As he races against time to unmask a mole within the agency, Ethan confronts his personal fears and the boundaries of the meticulous strategies he's known for. This slow-burn thriller follows Ethan's transformation in a world where every second could spell disaster, culminating in a showdown with Lena Grey\u2014a former ally who has turned the clockwork of espionage into her personal vendetta. Through grit and cunning, Ethan must adapt his methods, realizing that in the world of espionage, the most powerful weapon is a well-timed intuition.", "characters": [ { "name": "Ethan Cross", "role": "Protagonist", "description": "A meticulous and resourceful counter-intelligence operative known for his analytical mind and calm demeanor under pressure.", "arc": "Ethan transforms from a methodical planner to a decisive action-taker as he confronts his personal fears and realizes the importance of instinct." }, { "name": "Lena Grey", "role": "Antagonist", "description": "A brilliant but disillusioned former operative now turned mole, seeking vengeance against the agency she believes wronged her.", "arc": "Lena starts with a single-minded focus on revenge but gradually becomes conflicted as old loyalties resurface." }, { "name": "Dr. Julia Ward", "role": "Supporting Character", "description": "An astute psychologist who helps Ethan manage the stress of his demanding role and assists in profiling Lena's psychological state.", "arc": "Julia grows from a secondary advisory role to a key player in helping Ethan unearth Lena's motivations." }, { "name": "Michael Garner", "role": "Supporting Character", "description": "Ethan's trusted field partner and an expert in electronic surveillance, providing vital technical support.", "arc": "Michael's experience is tested as he learns to adapt to unpredictable situations, becoming more versatile in his approach." } ], "plotPoints": { "exposition": "Ethan Cross is tasked with investigating a series of leaked documents that could compromise national security. The leaks point to an insider within the agency.", "risingAction": "As Ethan dives deeper, he uncovers a trail leading to Lena Grey, a former colleague presumed dead. Evidence mounts as Ethan closes in, forcing him to question his long-standing methodologies.", "climax": "Ethan finally confronts Lena, who has rigged a trap to destroy critical evidence. In a tense standoff, Ethan must choose between following protocol or taking a risk to stop her.", "fallingAction": "With quick thinking and a new reliance on intuition, Ethan manages to disarm the trap. Lena, deflated, questions her own motives as old memories of camaraderie surface.", "resolution": "Lena is apprehended, the mole hunt ends, and Ethan reflects on his journey, acknowledging the balance between calculated strategy and spontaneity." }, "centralConflict": "Ethan Cross must identify and capture a mole within the agency who is leaking classified information, while dealing with his own rigid attachment to protocol in a dynamically evolving threat landscape.", "setting": "The story unfolds across various global locations including the bustling intelligence hub of Langley, a remote cabin in the Swiss Alps, and the teeming streets of Berlin, lending an authentic and international scope to the narrative.", "majorTurningPoints": [ "Ethan discovers the identity of the mole as his former colleague Lena Grey.", "Lena executes a series of diversions leading to a crisis within the agency.", "Ethan's adherence to protocol nearly costs him a critical breakthrough.", "Ethan's confrontation with Lena culminates in an uncharacteristic display of intuition that saves the mission." ], "characterArcs": { "protagonistArc": "Ethan evolves from strictly adhering to procedures to embracing a balance between strategy and instinctive decision-making, essential in high-stakes situations.", "antagonistArc": "Lena's journey from spite-fueled revenge to questioning her own motivations reflects a shift from isolation to an internal struggle with her past loyalties.", "supportingCharactersArcs": [ "Julia grows from providing psychological insights to playing an active role in strategizing the final approach to Lena.", "Michael transitions from a technical support role to becoming a crucial element in executing Ethan\u2019s plans, emphasizing adaptability." ] } } d Maari na naming gamitin muli ang schema na ito para makabuo ng maraming solusyon sa isang pipeline na may matitibay na garantiya tungkol sa mga field na nilalaman ng output. Konklusyon Nagbibigay-daan sa iyo ang structured meta-prompting na tukuyin ang structure sa mabilisang paraan, na ginagawang mas maaasahan ang mga output ng LLM para sa mga downstream na proseso. Manatiling nakatutok para sa susunod na post, kung saan tutuklasin namin ang pagsasama-sama ng structured meta-prompting sa iba pang mga diskarte.
