Revisiting RAG Retrievers: An Information Theoretic Benchmark
cs.IR, cs.AI, cs.LG
Submitted: 2026-02-25
Updated: 2026-11-02
Code: https://github.com/CapitalOne-Research/Migrasc
Terminology
Sources
- RePCS: Diagnosing Data Memorization in LLM-Powered Retrieval-Augmented Generation
- M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation
- Revisiting RAG Ensemble: A Theoretical and Mechanistic Analysis of Multi-RAG System Collaboration
- Influence Guided Context Selection for Effective Retrieval-Augmented Generation
- From Local to Global: A Graph RAG Approach to Query-Focused Summarization
- RAGBench: Explainable Benchmark for Retrieval-Augmented Generation Systems
- LightRAG: Simple and Fast Retrieval-Augmented Generation
- From RAG to Memory: Non-Parametric Continual Learning for Large Language Models
- Prompt-Response Semantic Divergence Metrics for Faithfulness Hallucination and Misalignment Detection in Large Language Models
- HyPA-RAG: A Hybrid Parameter Adaptive Retrieval-Augmented Generation System for AI Legal and Policy Applications
- CReSt: A Comprehensive Benchmark for Retrieval-Augmented Generation with Complex Reasoning over Structured Documents
- Attributing Response to Context: A Jensen-Shannon Divergence Driven Mechanistic Study of Context Attribution in Retrieval-Augmented Generation
- Operational Advice for Dense and Sparse Retrievers: HNSW, Flat, or Inverted Indexes?
- Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation
- Vendi-RAG: Adaptively Trading-Off Diversity And Quality Significantly Improves Retrieval Augmented Generation With LLMs
- Retrieval Augmentation Reduces Hallucination in Conversation
- BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models
- OmniEval: An Omnidirectional and Automatic RAG Evaluation Benchmark in Financial Domain
- BIRCO: A Benchmark of Information Retrieval Tasks with Complex Objectives
- ComposeRAG: A Modular and Composable RAG for Corpus-Grounded Multi-Hop Question Answering
Related papers
- Revisiting Text Ranking in Deep Research
- GateSID: One Gate to Balance Semantic and Collaborative Signals across the Item Lifecycle
- SAM-D2Q: Aligning Multimodal Doc2Query with Search Demand and Conversion for E-commerce
- GINGER: Spicing Up Graph-Based Vector Search
- Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA
- One-Step Retrieval Framework for Real-Time Sponsored Search Ads Using Hierarchical Text Representations