Explore
Tags
Browse articles across categories by shared concepts, technologies, formats, and other controlled classifications.
Prompt Design
Methods for writing, structuring, reusing, and improving instructions and examples supplied to an LLM.
Browse articlesRetrieval-Augmented Generation
Systems that retrieve external evidence and provide it to an LLM to support generated answers, including their pipeline, data lifecycle, and failure modes.
Browse articlesLLM Agents
LLM-driven systems that select actions or tools over one or more steps, including autonomy, state, budgets, and stopping behavior.
Browse articlesTool Calling
The interface and control flow through which an LLM proposes a tool action and an application validates, executes, and handles its result.
Browse articlesLLM Evaluation
Methods for measuring, testing, comparing, and interpreting LLM or LLM-application behavior against explicit criteria.
Browse articlesOutput Validation
Checking LLM-generated data or actions against explicit structural, semantic, evidence, or application requirements before use.
Browse articlesData Privacy
Managing sensitive information and disclosure risk when supplying data to LLM services or processing it in LLM applications.
Browse articlesLLM Security
Threats and defensive boundaries specific to LLM systems, including untrusted instructions, compromised evidence, and constrained actions.
Browse articlesHuman Review
Designing decisions and workflows that route uncertain, consequential, or otherwise unsuitable LLM outputs to a person for review.
Browse articlesLocal LLMs
Running or assessing language models on user-controlled or locally managed hardware rather than relying only on hosted inference.
Browse articlesMultimodal AI
LLM-based systems that process or produce combinations of text, images, audio, or other non-text modalities.
Browse articlesModel Training
Processes and objectives that shape model parameters before or during adaptation, including pretraining, instruction tuning, fine-tuning, and optimization.
Browse articlesModel Inference
How a trained model is used at runtime to process context and generate outputs, including decoding and inference resource behavior.
Browse articlesTokenization
How text is divided into model-specific tokens and how token counts affect model inputs, generation, limits, or cost.
Browse articlesContext Management
How the information available to an LLM during a request is assembled, budgeted, retained, summarized, or limited.
Browse articlesEmbeddings
Numerical representations used to compare text or other items, including their interpretation and role in retrieval.
Browse articlesSemantic Search
Finding and ranking information by meaning-related representations, including vector, hybrid, and approximate-neighbor retrieval methods.
Browse articlesLLM Deployment
Selecting, integrating, and operating model services and LLM application infrastructure under practical resource and service constraints.
Browse articlesLLM Use-Case Selection
Assessing whether and how an LLM is suitable for a task compared with simpler alternatives, using value, risk, evidence, and operational constraints.
Browse articles