Más contexto no es mejor. Paradoja de la dilución vectorial en RAG corporativos
Técnicas recientes de "Contextualized Chunking" inyectan resúmenes para mejorar el contexto en RAG, pero introducen una "dilución vectorial" que opaca el contenido local. Evaluando distintos ratios de inyección, demostramos una curva en "U invertida": una inyección moderada mejora el "Recall" (+18%), pero superar un umbral crítico (CIR > 0.4) reduce la precisión en un 22% para consultas específicas. Proponemos un marco teórico para calcular el ratio óptimo de inyección. -- Recent "Contextualized Chunking" techniques inject summaries to improve RAG context but introduce "vector dilution" drowning out local content. Evaluating various injection ratios, we demonstrate an "inverted U" curve: moderate injection boosts Recall (+18%), but exceeding a critical threshold (CIR > 0.4) drops precision by 22% for specific queries. We propose a theoretical framework to calculate the optimal injection ratio.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Detecting Contextomized Quotes in News Headlines by Contrastive Learning
Quotes are critical for establishing credibility in news articles. A direct quote enclosed in quotation marks has a strong visual appeal and is a sign of a reliable citation. Unfortunately, this journalistic practice is …
ArticlesContrastive LearningSemantic Preserving Embeddings for Generalized Graphs
A new approach to the study of Generalized Graphs as semantic data structures using machine learning techniques is presented. We show how vector representations maintaining semantic characteristics of the original data c…
BIG-bench Machine LearningEntity RetrievalRetrievalTécnicas Quantum-Inspired en Tensor Networks para Contextos Industriales
In this paper we present a study of the applicability and feasibility of quantum-inspired algorithms and techniques in tensor networks for industrial environments and contexts, with a compilation of the available literat…
Tensor NetworksMetric learning approach for graph-based label propagation
The efficiency of graph-based semi-supervised algorithms depends on the graph of instances on which they are applied. The instances are often in a vectorial form before a graph linking them is built. The construction of …
Metric LearningDecorrelated Vectorial Total Variation
This paper proposes a new vectorial total variation prior (VTV) for color images. Different from existing VTVs, our VTV, named the decorrelated vectorial total variation prior (D-VTV), measures the discrete gradients of …