Introduces CHARM framework that detects cascading hallucinations in agentic RAG at 89.4% rate with 5.3% false positives and reduces error propagation by 82.1% on multi-hop QA benchmarks.
hub Mixed citations
Ragas: Automated Evaluation of Retrieval Augmented Generation
Mixed citation behavior. Most common role is background (50%).
35
Pith papers citing it
Background
50% of classified citations
abstract
We introduce Ragas (Retrieval Augmented Generation Assessment), a framework for reference-free evaluation of Retrieval Augmented Generation (RAG) pipelines. RAG systems are composed of a retrieval and an LLM based generation module, and provide LLMs with knowledge from a reference textual database, which enables them to act as a natural language layer between a user and textual databases, reducing the risk of hallucinations. Evaluating RAG architectures is, however, challenging because there are several dimensions to consider: the ability of the retrieval system to identify relevant and focused context passages, the ability of the LLM to exploit such passages in a faithful way, or the quality of the generation itself. With Ragas, we put forward a suite of metrics which can be used to evaluate these different dimensions \textit{without having to rely on ground truth human annotations}. We posit that such a framework can crucially contribute to faster evaluation cycles of RAG architectures, which is especially important given the fast adoption of LLMs.
hub tools
citation-role summary
background 6
citation-polarity summary
roles
background 5representative citing papers
LogDx-CI benchmark shows hybrid grep+tail reducers achieve top diagnosis quality at low cost, agent loops shrink quality variance across reducers, and cross-family LLM summarizers outperform same-family pairs.
EnterpriseDocBench shows hybrid retrieval edges out BM25 and dense embeddings in end-to-end document pipelines, with weak inter-stage correlations and a gap between 85.5% factual accuracy and 0.40 average completeness.
GSAR is a grounding-evaluation framework for multi-agent LLMs that uses a four-way claim typology, evidence-weighted asymmetric scoring, and tiered recovery decisions to detect and mitigate hallucinations.
Long-horizon enterprise AI agents' decisions decompose into four measurable axes, with benchmark experiments on six memory architectures revealing distinct weaknesses and reversing a pre-registered prediction on summarization.
AgentProp-Bench shows substring judging agrees with humans at kappa=0.049, LLM ensemble at 0.432, bad-parameter injection propagates with ~0.62 probability, rejection and recovery are independent, and a runtime fix cuts hallucinations 23pp on GPT-4o-mini but not Gemini-2.0-Flash.
RAGognizer adds a detection head to LLMs for joint training on generation and token-level hallucination detection, yielding SOTA detection and fewer hallucinations in RAG while preserving output quality.
DotRAG reformulates graph retrieval as query-guided path reasoning with Division of Thought, reporting SOTA results on MetaQA and UltraDomain for multi-hop tasks.
StratRAG is a new benchmark dataset for multi-hop retrieval in RAG systems with noisy document pools, where hybrid retrieval reaches Recall@2 of 0.70 but bridge questions remain harder at 0.67.
MultiHop-RAG is a new benchmark dataset demonstrating that existing retrieval-augmented generation systems perform poorly on multi-hop queries requiring retrieval and reasoning over multiple evidence pieces.
Agentic hybrid RAG with a new muon collider benchmark outperforms baselines in retrieval effectiveness, answer quality, evidence coverage, and factual grounding.
Presents the Cross-Vendor Sola ISPM Benchmark and reports that adding relational context raises AI answer correctness by 34% and cuts exploration queries by 70% on multi-vendor identity tasks.
GroundedCache reduces unsafe-served rate in RAG answer caching to 0-1.5% (vs 15-51.5% naive) via four validation gates while keeping p50 latency within 1.07x of no-cache baseline.
For binary LLM judge validation, Pearson's r, Spearman's ρ, Kendall's τ_b, phi, and Matthews correlation all equal a single number on non-degenerate data, Cohen's κ supplies the extra signal on label-rate drift, and a reporting checklist is provided.
A framework using activation-based features from small open-weight proxy models detects LLM hallucinations with higher AUC than ReDeEP on RAGTruth, performing consistently across seven analyzer architectures.
CARE, a context-aware LLM judge, outperforms standard methods when evaluating multi-hop retrieval quality in RAG systems.
W-RAC decouples extraction from semantic planning via structured units and LLM grouping to match traditional retrieval performance at roughly 10x lower LLM token cost.
Introduces a 93-question multimodal RAG benchmark with phrase-level recall and embedding-based hallucination metrics, finding closed-source pipelines outperform open-source ones especially on cross-modal and cross-document tasks.
GraphRAG improves comprehensiveness and diversity of answers to global questions over million-token document sets by constructing entity graphs and hierarchical community summaries before combining partial responses.
ARMOR optimizes retrievers via joint RAG-likelihood and InfoNCE training with regularization toward the base encoder, yielding improved retrieval and QA on telecom benchmarks.
CRAFT introduces a query-conditioned pipeline with dynamic keyframe selection, ASR, and a hybrid critic loop that achieves top scores on MAGMaR 2026 for grounded multi-video question answering.
RAG-DIVE uses an LLM to dynamically generate, validate, and evaluate multi-turn dialogues for assessing RAG system performance in interactive settings.
IoDResearch is a private data-centric Deep Research framework that uses FAIR digital objects, atomic knowledge units, heterogeneous graph indexes, and a multi-agent system to outperform standard RAG baselines on retrieval, QA, and report generation tasks.
Proposes an intent-RAG framework that combines RAG, machine reasoning, and generative AI to interpret application intents and generate network intents, outperforming LLMs and vanilla RAG in translation tasks.
citing papers explorer
-
Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation
Introduces CHARM framework that detects cascading hallucinations in agentic RAG at 89.4% rate with 5.3% false positives and reduces error propagation by 82.1% on multi-hop QA benchmarks.
-
GSAR: Typed Grounding for Hallucination Detection and Recovery in Multi-Agent LLMs
GSAR is a grounding-evaluation framework for multi-agent LLMs that uses a four-way claim typology, evidence-weighted asymmetric scoring, and tiered recovery decisions to detect and mitigate hallucinations.
-
Four-Axis Decision Alignment for Long-Horizon Enterprise AI Agents
Long-horizon enterprise AI agents' decisions decompose into four measurable axes, with benchmark experiments on six memory architectures revealing distinct weaknesses and reversing a pre-registered prediction on summarization.
-
RAGognizer: Hallucination-Aware Fine-Tuning via Detection Head Integration
RAGognizer adds a detection head to LLMs for joint training on generation and token-level hallucination detection, yielding SOTA detection and fewer hallucinations in RAG while preserving output quality.
-
DOTRAG: Retrieval-Time Reasoning Along Paths
DotRAG reformulates graph retrieval as query-guided path reasoning with Division of Thought, reporting SOTA results on MetaQA and UltraDomain for multi-hop tasks.
-
Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis
Agentic hybrid RAG with a new muon collider benchmark outperforms baselines in retrieval effectiveness, answer quality, evidence coverage, and factual grounding.
-
Cross-Vendor Sola ISPM Benchmark: Evaluating Agentic AI for Federated Identity Security Reasoning
Presents the Cross-Vendor Sola ISPM Benchmark and reports that adding relational context raises AI answer correctness by 34% and cuts exploration queries by 70% on multi-vendor identity tasks.
-
Grounded Cache Routing for Retrieval-Augmented Generation: When Is It Safe to Reuse an Answer?
GroundedCache reduces unsafe-served rate in RAG answer caching to 0-1.5% (vs 15-51.5% naive) via four validation gates while keeping p50 latency within 1.07x of no-cache baseline.
-
Hallucination Detection via Activations of Open-Weight Proxy Analyzers
A framework using activation-based features from small open-weight proxy models detects LLM hallucinations with higher AUC than ReDeEP on RAGTruth, performing consistently across seven analyzer architectures.
-
Evaluating Multi-Hop Reasoning in RAG Systems: A Comparison of LLM-Based Retriever Evaluation Strategies
CARE, a context-aware LLM judge, outperforms standard methods when evaluating multi-hop retrieval quality in RAG systems.
-
Web Retrieval-Aware Chunking (W-RAC) for Efficient and Cost-Effective Retrieval-Augmented Generation Systems
W-RAC decouples extraction from semantic planning via structured units and LLM grouping to match traditional retrieval performance at roughly 10x lower LLM token cost.
-
FATHOMS-RAG: A Framework for the Assessment of Thinking and Observation in Multimodal Systems that use Retrieval Augmented Generation
Introduces a 93-question multimodal RAG benchmark with phrase-level recall and embedding-based hallucination metrics, finding closed-source pipelines outperform open-source ones especially on cross-modal and cross-document tasks.
-
From Local to Global: A Graph RAG Approach to Query-Focused Summarization
GraphRAG improves comprehensiveness and diversity of answers to global questions over million-token document sets by constructing entity graphs and hierarchical community summaries before combining partial responses.
-
ARMOR: Adaptive Retriever Optimization for Low-Resource Telecom Question Answering
ARMOR optimizes retrievers via joint RAG-likelihood and InfoNCE training with regularization toward the base encoder, yielding improved retrieval and QA on telecom benchmarks.
-
CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering
CRAFT introduces a query-conditioned pipeline with dynamic keyframe selection, ASR, and a hybrid critic loop that achieves top scores on MAGMaR 2026 for grounded multi-video question answering.
-
RAG-DIVE: A Dynamic Approach for Multi-Turn Dialogue Evaluation in Retrieval-Augmented Generation
RAG-DIVE uses an LLM to dynamically generate, validate, and evaluate multi-turn dialogues for assessing RAG system performance in interactive settings.
-
IoDResearch: Deep Research on Private Heterogeneous Data via the Internet of Data
IoDResearch is a private data-centric Deep Research framework that uses FAIR digital objects, atomic knowledge units, heterogeneous graph indexes, and a multi-agent system to outperform standard RAG baselines on retrieval, QA, and report generation tasks.
-
RAG-Enabled Intent Reasoning for Application-Network Interaction
Proposes an intent-RAG framework that combines RAG, machine reasoning, and generative AI to interpret application intents and generate network intents, outperforming LLMs and vanilla RAG in translation tasks.
-
BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning
BADGER is a new enterprise evaluation framework that adds LLM-assisted SQL component extraction and a Hybrid-EX metric validated on 150 human-annotated queries to existing text-to-SQL and agentic assessment methods.
-
Evaluating Factual Density in Multi-Source RAG: A Study in Medical AI Accuracy
Introduces FD* metric for factual density in RAG and shows it alone reaches 100% top-5 saturation of Cochrane evidence on HealthFC where cosine similarity does not.
-
Graph-Augmented Retrieval for Cross-Entity Financial Sentiment Analysis: A Comparative Study
Graph-RAG improves entity recall by 6.4% and answer relevancy by 11.7% over vector RAG on relational financial queries, with no loss in semantic similarity.
-
Deepchecks: Evaluating Retrieval-Augmented Generation (RAG)
Deepchecks is a new multi-faceted evaluation framework for RAG that incorporates root cause analysis and production monitoring to assess reliability, relevance, and user satisfaction.
-
LLM-Oriented Information Retrieval: A Denoising-First Perspective
Argues for a denoising-first paradigm in LLM-oriented information retrieval, framing challenges via a four-stage progression and providing a taxonomy of signal-to-noise optimization techniques across the pipeline.
-
Beyond the Parameters: A Technical Survey of Contextual Enrichment in Large Language Models: From In-Context Prompting to Causal Retrieval-Augmented Generation
The survey unifies LLM augmentation techniques along the single axis of structured context supplied at inference time and supplies a literature screening protocol plus deployment decision framework.
-
Retrieval-Augmented Generation for Large Language Models: A Survey
A survey of RAG paradigms, components, benchmarks, and challenges for improving LLMs on knowledge-intensive tasks.
-
Development of a Retrieval-Augmented Generation Virtual Assistant for Enhanced Information Discovery at Rubin Observatory
Prototype RAG virtual assistant integrates Rubin Observatory documentation using Weaviate, LangChain, and GPT for conversational semantic search.
-
A Survey on Retrieval-Augmented Text Generation for Large Language Models
A survey that categorizes RAG methods for LLMs into four retrieval-centric stages, reviews their evolution and evaluation, and outlines challenges and future directions.