More in AI & ML — page 39
Mode Collapse: When Your AI Gets Stuck in a Rut
Mode collapse is when a generative AI finds a “cheat” and produces the same few outputs over and over. This is a classic failure in GANs where the generator stops learning the full data distribution.
Text-to-Image Synthesis: From Prompt to Picture
Text-to-image models translate words into pixels by learning statistical links between text and images. They power creative tools like DALL-E but don't truly understand prompts, leading to errors in logic like counting or spatial arrangement.
Generative Inpainting: Filling in the Blanks with AI
Generative inpainting is like Photoshop's 'Content-Aware Fill' on steroids. It uses AI to reconstruct missing or unwanted parts of an image, perfect for removing objects or repairing old photos.
Hierarchical AI Agents: The Org Chart for AI
Think of a corporate org chart for AI. A top-level agent breaks a big goal into smaller tasks and delegates them to specialized, lower-level agents. This is used for complex problems like automating software development. The main risk is coordination overhead.
Multi-Agent Systems: A Team of AIs, Not a Monolith
Instead of one giant AI, a Multi-Agent System is a team of specialized AIs that collaborate on a complex problem. This appears in LLM-powered coding assistants and complex simulations.
Task Decomposition: Teaching LLMs to Plan
Task decomposition for an LLM agent is like writing a recipe: break a big goal into a checklist of small, executable steps. It's vital for complex requests like planning a trip, but a bad initial plan can cause cascading failures that doom the entire process.
LLM Agents: Giving Models Tools and a Plan
An LLM Agent gives a model tools and a plan to solve complex problems. Instead of just answering a question, it breaks it down, uses APIs or code interpreters, and remembers past steps. The footgun is thinking RAG is an agent; agents need planning and.
Graph RAG: Answering Questions with Connected Facts
Graph RAG answers complex questions by exploring a map of connected facts (a knowledge graph) instead of just searching flat text. Use it for queries needing synthesis, like finding drugs for a disease made by companies in a specific country.
HyDE: Find Documents by Embedding a Fake Answer
Instead of embedding a short query, HyDE uses an LLM to generate a full, hypothetical answer document. This vector, representing an ideal answer, is then used to find similar real documents, improving zero-shot retrieval. The generated document is fictional.
Self-Querying Retriever: Let an LLM Write Its Own Filters
A self-querying retriever uses an LLM to turn a natural language question into a structured query with metadata filters. It lets users ask things like "Find documents about Python from before 2020," which a simple vector search can't do.
LLMs Get 'Lost in the Middle' of Long Contexts
LLMs struggle to find information buried in the middle of long prompts. Performance is highest when key facts are at the beginning or end of the context. This impacts multi-document QA and RAG.
Context Stuffing: Giving LLMs Short-Term Memory
Context stuffing adds external documents to an LLM's prompt, giving it temporary, task-specific knowledge. Use it for one-off questions on specific docs, but beware: it fails when documents exceed the model's context window limit, causing truncated data.
Hybrid Search: Combining Keyword and Vector Search
Hybrid search combines keyword precision with vector search's conceptual understanding in one query. It excels at retrieving relevant documents for RAG by finding both exact matches (like names) and similar ideas.
Dense Passage Retrieval (DPR): Semantic Search for QA
DPR finds answers by meaning, not just keywords. It converts questions and documents into vectors and finds the closest matches, forming the core of Retrieval-Augmented Generation (RAG).
Embedding Models: The 'Retrieval' in RAG
An embedding model acts like a librarian for your data, converting text into numerical vectors so similar concepts are grouped together. In RAG, it finds relevant documents to feed an LLM, but using the wrong model type will yield poor retrieval results.
QLoRA: Finetune Huge LLMs on a Single GPU
QLoRA lets you finetune massive LLMs on one GPU by freezing the model in a 4-bit state and only training tiny adapter layers. Use it to adapt a 65B model with limited hardware. The footgun: performance hinges on high-quality data, not just the technique.
Proximal Policy Optimization (PPO): Stable RL Updates
PPO prevents destructive updates in reinforcement learning by "clipping" how much a policy can change at once, like a governor on an engine. It's a default for training LLMs with RLHF or robotics agents where stability is key.
Adapter Modules: Efficient LLM Fine-Tuning
Adapters are small modules plugged into a frozen LLM to avoid costly full fine-tuning. This lets you specialize a base model for many tasks by training tiny, swappable plugins instead of duplicating the entire model for each task.
Reward Modeling: Teaching an LLM What 'Good' Means
A reward model is a judge that scores an LLM's outputs based on human preferences. It learns to assign a numerical 'goodness' score to text, turning subjective quality into an optimizable signal for training models like ChatGPT.
Instruction Fine-Tuning: Teaching LLMs to Follow Orders
Instruction fine-tuning teaches a base LLM to follow commands, not just predict the next word. It turns a raw text-completion engine into a helpful assistant, enabling it to answer questions or summarize text. The footgun: it learns style, not facts.