Modelos de lenguaje grandes (LLMs)
Definición
Los grandes modelos de lenguaje son modelos basados en transformers entrenados con datos textuales masivos (y a veces multimodales). They exhibit emergent abilities: few-shot learning, razonamiento, and tool use when scaled and aligned (por ej. via RLHF).
Un modelo mental útil: preentrenamiento aprende predicción del siguiente token en enormes corpus y da al modelo amplio conocimiento y lenguaje ability. Instruction tuning (and similar) trains the model to follow user instructions and formats. Alignment (por ej. RLHF, DPO) shapes behavior to be helpful, honest, and safe. At inference time you can use the model zero-shot, few-shot, or augment it with recuperación (RAG) or tools (agents).
Cómo funciona
Pretraining learns predicción del siguiente token on large corpora and produce a base model. Optional fine-tuning (por ej. fine-tuning) adapts it to tasks or instruction formats; alignment (por ej. RLHF, DPO) optimizes human preference and safety. The deployed model is then used at inference time. You can call it zero-shot (no examples), few-shot (with prompt engineering), or augment it with RAG (recuperación as context) or agents (tools and loops). The diagram summarizes the training pipeline and the two main inference augmentations.
Casos de uso
LLMs are used wherever you need flexible language understanding or generation, from chat to code to analysis.
- Chat, summarization, and translation
- Code assistance and generation
- Question answering and research assistance (often with RAG or tools)
Ventajas y desventajas
| Pros | Cons |
|---|---|
| Flexible, one model for many tasks | Cost and latency |
| Strong few-shot performance | Hallucination, bias |
| Enables agents and tool use | Requires careful evaluation |
Documentación externa
- OpenAI – Models overview — GPT and capabilities
- Google AI for Developers — Gemini and APIs
- Anthropic – Models — Claude and documentation
- Hugging Face – NLP course — From transformers to LLMs