'dpo' 검색 결과

Train language models with SimPO using pairs of preferred and rejected answers. Configure preference training without a separate reference model, adapt settings for base or instruction models, and address unstable training, memory limits, or lost capabilities.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Training a base model with preferred and rejected answers
• Further tuning an instruction model with preference data
• Adjusting training settings for math and code tasks

Plan and configure large language model training with OpenRLHF to better match human preferences. Covers training method selection, distributed GPU setup, inference acceleration with vLLM, and troubleshooting memory or training stability issues.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Choosing a method for training models on human preferences
• Setting up PPO training with a reward model
• Configuring GRPO training without a critic model

Train or fine-tune language and vision models on Hugging Face Jobs without a local GPU. Covers dataset checks, hardware and cost planning, training monitoring, Hub storage, and GGUF conversion for local use.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Fine-tune a language model without a local GPU
• Validate preference data before DPO training
• Estimate training costs and choose GPU hardware

Generate ASP.NET Core pages, components, or API endpoints for creating, reading, updating, and deleting database records. Follow the project's existing styling and code conventions, with database migrations and verification of the generated functionality.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Add database-backed record management pages
• Generate Minimal API endpoints for a model
• Create Blazor forms that match existing styling

Debug 404 errors for API endpoints in microservice architectures where code exists but routing sends requests to wrong service. Use when: (1) API endpoint returns 404 despite code being deployed, (2) Multiple services share same hostname via HTTPRoute/Ingress, (3) Service logs show "Health server" but you expected API endpoints, (4) Monorepo deploys multiple images and endpoint might be in different service than expected. Covers: verifying which service handles which paths, HTTPRoute path precedence, testing endpoints directly from inside cluster, matching image tags across services.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

API Endpoint Builder workflow skill. Use this skill when the user needs Builds production-ready REST API endpoints with validation, error handling, authentication, and documentation. Follows best practices for security and scalability and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

arkcli +deploy：明确指定型号创建接入点也由本 Skill 接管，包括 Retiring 型号；推荐过滤不替代指定目标的实际创建结果。普通创建推理接入点（Endpoint）的统一首选入口。用户说『创建/新建/create 一个 endpoint/接入点』或『部署/上线/deploy 某模型』时优先走这里；**但脚本化 / CI / 无护栏 / 原始 raw CRUD 创建是唯一例外，必须改走 `arkcli-infer-endpoint`，不能由本 skill 截获**。只有不命中该例外的普通创建，才在模型尚未选择、只给品牌/家族名，或当前轮只查询候选时进入本 skill 完成模型澄清；候选要用实时 search 返回的 `name` 与 `primary_version` 组合成可直接传给 `--model` 的完整 ID。对**已有** Endpoint 做获取/列表/启停/更新等全生命周期管理也走 arkcli-infer-endpoint；本 skill 只负责带产品护栏的一键创建。创建成功后会自动把多语言调用示例渲染到 ./ark-examples/<ep-id>/。反触发：TTS/ASR/语音模型不能 +deploy，只能转 models search 说明广场可搜但 arkcli 不支持 Endpoint 创建。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Create a PostHog endpoint with the right shape on the first try — covers query kind choice, name conventions, what to expose as variables (HogQL code_name vs insight breakdown), data_freshness_seconds, and whether to materialise on day one. Use when the user says "create an endpoint", "expose this query as an API", "turn this insight into an endpoint", or asks for help structuring a new endpoint. Steers away from common mistakes: materialising a query with cohort breakdowns or compare mode, inline-only variables on a materialised endpoint, unbounded date ranges, ambiguous names.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Expert knowledge of REST endpoint patterns in melosys-api's frontend-api module. Use when: (1) Creating new REST endpoints for melosys-web, (2) Understanding controller patterns and Swagger documentation, (3) Working with DTOs and request/response objects, (4) Understanding exception handling and HTTP status codes, (5) Adding access control to endpoints. Triggers: "new endpoint", "REST controller", "@RestController", "add a DTO", "ResponseEntity", "Aksesskontroll", "access control on endpoint", "Swagger/@Operation", "return 404/403/401/400", "@Protected", "frontend-api", "search endpoint", "/sok".

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hf 명령어로 Hugging Face Hub의 모델과 데이터셋을 찾고 다운로드·업로드합니다. 저장소, 실행 환경과 로컬 캐시 관리도 다룹니다.

• 모델과 데이터셋을 찾아 다운로드하기
• 로컬 파일을 Hub에 업로드하기
• 저장소 생성과 브랜치·태그 관리하기

Turn a system description into REST API endpoints, the URLs and operations applications use to exchange data. Choose an endpoint list or detailed specifications covering authentication, request fields, responses, error codes, and pagination for lists.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Listing the API endpoints a new system needs
• Drafting request and response specifications
• Designing authentication and error responses

Build ASP.NET Minimal API endpoints with explicit request and response types, input validation, and standard error responses. Organize related endpoints and add OpenAPI documentation, a structured description of API operations, parameters, and schemas.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Building Minimal API endpoints with explicit types
• Adding operation and parameter descriptions to OpenAPI
• Organizing related endpoints with MapGroup()

Guide language model training with TRL, from learning to follow instructions to matching human preferences. Covers method selection, training data, reward models that score responses, evaluation, and adjustments for memory limits or unstable training.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Train a model on instruction and response examples
• Align responses using preferred and rejected answer pairs
• Build a training pipeline using human feedback

Audit whether an AI agent follows its intended purpose, business rules, and permission boundaries. The skill guides setup discovery, test configuration, execution against an agent endpoint or substitute model, and interpretation of scores from selected judging models.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Auditing an agent's compliance with business rules
• Testing permission boundaries and dangerous tools
• Evaluating an agent through its HTTP endpoint

Plan a prospective clinical study by ranking endpoints, estimating participant numbers and statistical power, and assessing feasibility. Outputs include assumptions and named reviewers, providing material for expert review before a protocol is finalized.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Choosing primary and secondary study endpoints
• Estimating sample size with dropout allowance
• Assessing feasibility before a study review

Configure OpenTelemetry to send AI application traces, records of execution steps, to Confident AI. Reuse existing exporters, choose the regional endpoint, and add AI-specific attributes without requiring the deepeval Python package.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Send existing AI application traces to Confident AI
• Configure tracing for a non-Python AI application
• Choose the US or EU endpoint for an API key

Guide language model training with TRL, from learning to follow instructions to matching human preferences. Covers method selection, training data, reward models that score responses, evaluation, and adjustments for memory limits or unstable training.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Train a model on instruction and response examples
• Align responses using preferred and rejected answer pairs
• Build a training pipeline using human feedback

Build serverless endpoints that relay external API requests for a dashboard while keeping API keys on the server. Includes Vercel examples for input checks, caching, cross-origin access, and consistent JSON responses.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Connect a dashboard to external APIs without exposing API keys
• Create endpoints for stock quotes and news
• Standardize API responses and error handling

Build or update ASP.NET Core API endpoints with defined request and response types, appropriate HTTP status codes, OpenAPI documentation, and centralized error handling. Also creates .http request files for manually checking endpoint behavior.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Add API endpoints following existing project conventions
• Document API inputs and responses with OpenAPI
• Standardize error responses and exception handling

API Endpoint Builder workflow skill. Use this skill when the user needs Builds production-ready REST API endpoints with validation, error handling, authentication, and documentation. Follows best practices for security and scalability and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

API endpoint design and testing for vehicle insurance data platform. Use when designing new API endpoints, testing existing ones, validating response formats, or debugging API issues. Covers 11 core endpoints including 3 new pie chart distribution endpoints, parameter validation, error handling, and integration patterns.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Deep API security assessment beyond surface scanning. Covers the full OWASP API Security Top 10 (2023): Broken Object Level Authorization (BOLA / IDOR), Broken Authentication, Broken Object Property Level Authorization (mass assignment + excessive data exposure), Unrestricted Resource Consumption, Broken Function Level Authorization (BFLA / vertical privilege escalation), Unrestricted Access to Sensitive Business Flows, Server-Side Request Forgery via API parameters, Security Misconfiguration, Improper Inventory Management (shadow/zombie/deprecated endpoints, v1/v2 drift), and Unsafe Consumption of third-party APIs. Works across REST, GraphQL, gRPC, SOAP, and MCP servers. Discovers APIs from OpenAPI/Swagger specs, GraphQL introspection, gRPC reflection, .well-known endpoints, JS bundles, and traffic capture. Uses kiterunner, ffuf, schemathesis, restler-fuzzer, openapi-fuzzer, graphql-cop, clairvoyance, batchql, inql, jwt_tool, postman, mitmproxy, and manual http(action="request", ...) payloads. Every technique includes actual payloads, commands, and verification logic. Chains from /pentester or /codebase when API endpoints are discovered, chains into /web-exploit when classic injection points are found in API parameters, chains into /post-exploit when RCE is achieved, and chains into /ai-redteam when an LLM/AI endpoint is discovered (chat APIs, completion endpoints, RAG search, agentic tool-use, MCP servers).

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Manages Agent Platform serving endpoints. Use when you need to create, list, describe, update, or delete serving endpoints for model deployment on Agent Platform. Also use when troubleshooting endpoint permission, quota, or resource busy errors. Don't use for deploying models to endpoints or for running model evaluations.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Deploy open models or custom weights from Model Garden to Agent Platform endpoints, check the status of an in-progress deployment operation, or clean up resources by undeploying models and deleting endpoints. Use when asked to actively deploy a model, list the Model Garden CATALOG of available models, check if a specific model is deployable (`gcloud ai model-garden models list-deployment-config`), query deployment cost, troubleshoot deployment errors (like quota limits), or undeploy/clean up endpoints. Also use when copying and deploying a 1P Tuned Model. Don't use for pure listing/discovery questions of the form "is X deployed?", "list my endpoints", or "which regions have models running?" — for those use `agent-platform-endpoint-management`. Don't use for public Vertex AI deployments (use the `vertex-deploy` skill) or for running model evaluations (use the `agent-platform-eval-flywheel` skill).

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.