mlsanityData to 7 Oct 2026

To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG

arXiv 2606.25191 ↗First posted 23 Jun 2026cs.AI cs.CLcited by 2

On the map: Retrieval-augmented generation → Agentic RAG, agents and memory

Abstract (arXiv)

Multi-agent document assessment for retrieval-augmented generation is computationally expensive, driving practitioners toward smaller, deployable models whose assessment mechanisms remain poorly understood. We conduct a controlled study of training-free interventions on 7B-9B instruction-tuned models across diverse QA benchmarks, revealing a sharp dichotomy in how models benefit from assessment. For weaker baselines, the dominant mechanism is per-document isolation. Astoundingly, assessment-free isolation matches full multi-agent assessment, demonstrating that resolving multi-document context confusion, rather than scoring quality, drives outsized gains of up to 50 percentage points. Conversely, for strong baselines where scoring quality matters, we introduce Reasoning-Score Coupling, a label-free perturbation probe that classifies scoring behavior. Integrating these findings, we propose MADARA, a model-adaptive routing architecture. Crucially, MADARA's diagnostic thresholds derived from a single pilot model generalize zero-shot to four unseen model families, providing a robust, lightweight pipeline to eliminate computational overhead.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Agentic RAG, agents and memory, one of 15 directions in Retrieval-augmented generation. Builds on 38 of its 51 references that are in the map. First posted on arXiv 23 Jun 2026.

Builds on

38 of 51 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG15 Jan
  2. 2024MAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation31 Dec

Other references · 36

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  3. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  4. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  5. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,928
  6. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,800
  7. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  8. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,362
  9. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,203
  10. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  11. 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingcited by 1,081
  12. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 995
24 more
  1. 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 778
  2. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 633
  3. 2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 480
  4. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexitycited by 437
  5. 2023Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflictscited by 333
  6. 2021Do Prompt-Based Models Really Understand the Meaning of their Prompts?cited by 324
  7. 2024Knowledge Conflicts for LLMs: A Surveycited by 274
  8. 2024A Comprehensive Survey of Small Language Models in the Era of Large Language Models: Techniques, Enhancements, Applications, Collaboration with LLMs, and Trustworthinesscited by 125
  9. 2025Context Length Alone Hurts LLM Performance Despite Perfect Retrievalcited by 118
  10. 2024Small Language Models: Survey, Measurements, and Insightscited by 109
  11. 2023Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMscited by 103
  12. 2024Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreementcited by 99
  13. 2025Retrieval-Augmented Generation with Conflicting Evidencecited by 60
  14. 2025Measuring Chain of Thought Faithfulness by Unlearning Reasoning Stepscited by 43
  15. 2025Removal of Hallucination on Hallucination: Debate-Augmented RAGcited by 29
  16. 2025Free-MAD: Consensus-Free Multi-Agent Debatecited by 27
  17. 2025DRAGged into Conflicts: Detecting and Addressing Conflicting Sources in Search-Augmented LLMscited by 21
  18. 2025U-NIAH: Unified RAG and LLM Evaluation for Long Context Needle-In-A-Haystackcited by 20
  19. 2026Demystifying Multi-Agent Debate: The Role of Confidence and Diversitycited by 17
  20. 2025Beyond Consensus: Mitigating the Agreeableness Bias in LLM Judge Evaluationscited by 16
  21. 2025TruthfulRAG: Resolving Factual-level Conflicts in Retrieval-Augmented Generation with Knowledge Graphscited by 12
  22. 2025Robust Answers, Fragile Logic: Probing the Decoupling Hypothesis in LLM Reasoningcited by 11
  23. 2025Edge Deployment of Small Language Models, a comprehensive comparison of CPU, GPU and NPU backendscited by 4
  24. 2025Zero-RAG: Towards Retrieval-Augmented Generation with Zero Redundant Knowledgecited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026The Hallucination Signal Is a Mean Shift: Why Simple Probes Sufficecited by 0
  2. 2026Where Activation Sparsity and KV-Cache Sparsity Cross in LLM Decodingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Beyond Component Strength: Synergistic Integration and Adaptive Calibration in Multi-Agent RAG Systems
  2. 2026MADRAG: Multi-Agent Debate with Retrieval-Augmented Generation for Training-Free Analytic Essay Scoring
  3. 2026Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval
  4. 2026GRADRAG: Cross-Component Prompt Adaptation for Coordinated Multi-Agent RAG
  5. 2026Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model
  6. 2026A-RAG: Scaling Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces
  7. 2025SIRAG: Towards Stable and Interpretable RAG with A Process-Supervised Multi-Agent Framework
  8. 2025MA-RAG: Multi-Agent Retrieval-Augmented Generation via Collaborative Chain-of-Thought Reasoning