Fonti

Fonti autorevoli

Le definizioni, le tecniche e le pratiche di questo pack sono fondate sulle seguenti fonti primarie e ampiamente riconosciute: documentazione ufficiale dei fornitori di modelli, articoli di ricerca che hanno introdotto le tecniche chiave e standard di sicurezza della comunità. Dove una tecnica ha un articolo originario canonico, esso è citato insieme alla guida ufficiale mantenuta dal fornitore.

Documentazione dei fornitori di modelli

  • Anthropic - Panoramica sul prompt engineering (Claude Docs). https://docs.anthropic.com/en/docs/build-with-claude/prompt-engineering/overview
  • Anthropic - Usare i tag XML per strutturare i prompt. https://docs.anthropic.com/en/docs/build-with-claude/prompt-engineering/use-xml-tags
  • Anthropic - Pensiero esteso (extended thinking). https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking
  • Anthropic - Caching dei prompt. https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching
  • Anthropic - Uso di strumenti (function calling). https://docs.anthropic.com/en/docs/build-with-claude/tool-use/overview
  • Anthropic - Ridurre le allucinazioni. https://docs.anthropic.com/en/docs/test-and-evaluate/strengthen-guardrails/reduce-hallucinations
  • Anthropic - Costruire agenti efficaci. https://www.anthropic.com/engineering/building-effective-agents
  • Anthropic - Ingegneria del contesto efficace per agenti AI. https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents
  • OpenAI - Guida al prompt engineering (documentazione API). https://platform.openai.com/docs/guides/prompt-engineering
  • OpenAI - Structured Outputs. https://platform.openai.com/docs/guides/structured-outputs
  • Google - Strategie di prompting dell'API Gemini. https://ai.google.dev/gemini-api/docs/prompting-strategies

Articoli di ricerca fondamentali

  • Brown et al. (2020) - Language Models are Few-Shot Learners (GPT-3). https://arxiv.org/abs/2005.14165
  • Wei et al. (2022) - Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. https://arxiv.org/abs/2201.11903
  • Kojima et al. (2022) - Large Language Models are Zero-Shot Reasoners ("Let's think step by step"). https://arxiv.org/abs/2205.11916
  • Wang et al. (2022) - Self-Consistency Improves Chain of Thought Reasoning. https://arxiv.org/abs/2203.11171
  • Yao et al. (2022) - ReAct: Synergizing Reasoning and Acting in Language Models. https://arxiv.org/abs/2210.03629
  • Lewis et al. (2020) - Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. https://arxiv.org/abs/2005.11401
  • Bai et al. (2022) - Constitutional AI: Harmlessness from AI Feedback. https://arxiv.org/abs/2212.08073
  • Zheng et al. (2023) - Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena. https://arxiv.org/abs/2306.05685

Guide e standard della comunità

  • DAIR.AI - Prompt Engineering Guide. https://www.promptingguide.ai/
  • OWASP - Top 10 per le applicazioni con Large Language Model. https://owasp.org/www-project-top-10-for-large-language-model-applications/
  • Model Context Protocol - Specifica e documentazione. https://modelcontextprotocol.io/
  • RAGAS - Framework di valutazione per pipeline RAG (documentazione). https://docs.ragas.io/