Results for "meaning-based retrieval"

Breaking documents into pieces for retrieval; chunk size/overlap strongly affect RAG quality.

Prompt augmented with retrieved documents.

Retrieval based on embedding similarity rather than keyword overlap, capturing paraphrases and related concepts.

Architecture that retrieves relevant documents (e.g., from a vector DB) and conditions generation on them to reduce hallucinations.

A datastore optimized for similarity search over embeddings, enabling semantic retrieval at scale.

Enables external computation or lookup.

Coordinating tools, models, and steps (retrieval, calls, validation) to deliver reliable end-to-end behavior.

A measure of a model class’s expressive capacity based on its ability to shatter datasets.

AI subfield dealing with understanding and generating human language, including syntax, semantics, and pragmatics.

Optimal pathfinding algorithm.

Measures how one probability distribution diverges from another.

Quantifies shared information between random variables.

Optimization problems where any local minimum is global.

Temporal and pitch characteristics of speech.

Extending agents with long-term memory stores.

AI-assisted review of legal documents.

Harmonic mean of precision and recall; useful when balancing false positives/negatives matters.

Constraining outputs to retrieved or provided sources, often with citation, to improve factual reliability.

Mechanisms for retaining context across turns/sessions: scratchpads, vector memories, structured stores.

Structured graph encoding facts as entity–relation–entity triples.

Central catalog of deployed and experimental models.

Agents communicate via shared state.

Internal representation of environment layout.

AI supporting legal research, drafting, and analysis.

Requirement to preserve relevant data.

RL using learned or known environment models.

Exact likelihood generative models using invertible transforms.

Combines value estimation (critic) with policy learning (actor).

Learns the score (∇ log p(x)) for generative sampling.

Simple agent responding directly to inputs.