Namespace LMKit.Retrieval

DocumentIndexingResult: Describes the outcome of loading a document into PdfChat, including how the document was processed and its resource consumption.

DocumentQueryResult: Contains the result of a document query, including the generated response and references to source passages.

DocumentRag: Provides document-centric Retrieval-Augmented Generation (RAG) capabilities with built-in support for multi-page document processing, OCR, and vision-based document understanding.

DocumentRag.DocumentMetadata: Represents metadata associated with a document during import into a DocumentRag instance.

DocumentReference: Represents a reference to a specific location within a document, retrieved during a retrieval operation.

HtmlChunking: Provides HTML-aware chunking for retrieval workflows. The splitter uses AngleSharp to parse the HTML DOM and respects structural boundaries such as headings, sections, tables, and block-level elements to produce semantically coherent chunks.

HybridRetrievalStrategy: A retrieval strategy that combines vector (semantic) search and BM25 (keyword) search, fusing the results with weighted Reciprocal Rank Fusion (RRF) for improved recall.

HydeOptions: Configures how hypothetical answers are generated during HypotheticalAnswer retrieval.

MarkdownChunking: Provides Markdown-aware chunking configuration for retrieval workflows. The underlying splitter favors Markdown structural boundaries (such as headings and paragraph breaks) to produce chunks that preserve semantic coherence.

MultiQueryOptions: Configures how multiple query variants are generated and searched during MultiQuery retrieval.

PartitionSimilarity: Represents the result of a retrieval operation, capturing the similarity between a partition (or vector entry) and a target item.

PdfChat: Provides conversational question-answering over PDF documents by combining intelligent document understanding with grounded response generation.

QueryContextualizationOptions: Configures how follow-up questions are reformulated into self-contained queries during Contextual retrieval.

RagChat: Provides multi-turn conversational question-answering over a user-managed RagEngine, combining retrieval, query contextualization, and grounded response generation in a single turnkey interface.

RagEngine: Provides core functionalities for Retrieval-Augmented Generation (RAG) within a data processing system.

RagEngine.RagReranker: Encapsulates a reranking model and blending factor for adjusting raw similarity scores in RAG workflows.

RagQueryResult: Contains the result of a RAG query, including the generated response and the retrieved partitions used as context.

RetrievalDefaults: Provides default values shared across the retrieval subsystem.

TextChunking: Implements a recursive chunking strategy for partitioning text into manageable segments, known as "chunks," to support retrieval-augmented generation tasks.
This approach is particularly effective for processing extensive texts, systematically breaking them down into smaller segments that are easier to handle.
Unlike linear chunking methods that sequentially divide text, this recursive strategy dynamically adjusts the segmentation process based on the complexity and structure of the text.
This allows for more nuanced and efficient handling of text data, especially when dealing with nested or hierarchical information.

VectorRetrievalStrategy: A retrieval strategy that scores partitions using cosine similarity between the query embedding and partition embeddings.

VectorSearch: Provides methods for searching partitions across one or more data sources by comparing vector embeddings for similarity.

IChunking: Defines configurable settings for text chunking. Implementations control how input text is partitioned into chunks suitable for retrieval and embedding workflows.

IRetrievalStrategy: Defines a strategy for retrieving matching partitions from indexed data sources.

DocumentIndexingResult.DocumentIndexingMode: Specifies how a document was processed for retrieval.

PageProcessingMode: Specifies how document pages are interpreted during document preparation.

QueryGenerationMode: Specifies how retrieval queries are generated from user input before semantic search.