mlsanityData to 14 Oct 2026

A corrective agentic hybrid RAG and an operations-grounded evaluation for a scientific facility

arXiv 2607.24663 ↗First posted 27 Jul 2026physics.acc-ph cs.AI cs.IRcited by 0

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

also in 1 more
Abstract (arXiv)

Scientific user facilities accumulate decades of operational knowledge that no single search index covers: electronic logbooks, technical documents, internal wikis, operations chat messages, maintenance records, and live control-system data. We present APS-RAG, Advanced Photon Source Retrieval Augmented Generation, a deployed platform that makes the institutional knowledge at the Advanced Photon Source (APS) accessible to staff through natural-language queries, along with an operations-grounded evaluation. The retrieval engine fuses dense, sparse, and knowledge-graph (KG) channels with query-type-adaptive reciprocal-rank fusion, adds a corrective agentic loop, and runs a native-tool ReAct executor over a Model Context Protocol (MCP) tooling layer. We construct APS-Bench, a 50-question, question-answering (QA) dataset with auditable gold answers. Every retrieval-augmented variant numerically improves strict vital-nugget recall over a naive BM25 baseline (63.8%), with the full corrective Agentic GraphRAG scoring (70.3%). The cross-encoder reranker contributes significantly to answer quality: removing it and allowing the LLM to score relevance drastically reduces strict vital recall by 32.8%. The graph channel and corrective loop contribute positively as expected, but the performance gains are marginal. Additionally, we also compare the performance of open-source and closed-source LLMs in final answer synthesis. We release the APS-Bench construction methodology, the six-layer evaluation harness, and the underlying codebase, along with the '/aps-rag' retrieval agent skill framework, to support reproduction and adoption at other facilities. Together, the deployed platform and its operations-grounded evaluation present a promising workflow for trustworthy, statistically grounded AI assistance in facility operations, transferable to other large scientific instruments.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Agentic RAG, agents and memory, one of 15 directions in Retrieval-augmented generation. Builds on 39 of its 84 references that are in the map. First posted on arXiv 27 Jul 2026.

Builds on

39 of 84 references.

Closest ancestors · 2

Same direction, newest first

  1. 2026Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis9 Jun
  2. 2025Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG15 Jan

Other references · 37

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  5. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  6. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  7. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  8. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,212
  9. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  10. 2020MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformerscited by 1,463
  11. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,453
  12. 2024A Survey on LLM-as-a-Judgecited by 1,423
25 more
  1. 2025Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Modelscited by 1,261
  2. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,253
  3. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,073
  4. 2024LLM Evaluators Recognize and Favor Their Own Generationscited by 686
  5. 2019Passage Re-ranking with BERTcited by 645
  6. 2024RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrievalcited by 532
  7. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexitycited by 451
  8. 2023Ragas: Automated Evaluation of Retrieval Augmented Generationcited by 418
  9. 2024Graph Retrieval-Augmented Generation: A Surveycited by 324
  10. 2024LightRAG: Simple and Fast Retrieval-Augmented Generationcited by 315
  11. 2024Corrective Retrieval Augmented Generationcited by 197
  12. 2022An Analysis of Fusion Functions for Hybrid Retrievalcited by 51
  13. 2025When to use Graphs in RAG: A Comprehensive Analysis for Graph Retrieval-Augmented Generationcited by 48
  14. 2024Initial Nugget Evaluation Results for the TREC 2024 RAG Track with the AutoNuggetizer Frameworkcited by 28
  15. 2023Opportunities for Retrieval and Tool Augmented Large Language Models in Scientific Facilitiescited by 27
  16. 2025jina-reranker-v3: Last but Not Late Interaction for Listwise Document Rerankingcited by 24
  17. 2025Benchmarking LLM Faithfulness in RAG with Evolving Leaderboardscited by 20
  18. 2024Large Language Models for Human-Machine Collaborative Particle Accelerator Tuning through Natural Languagecited by 9
  19. 2024GAIA: A General AI Assistant for Intelligent Accelerator Operationscited by 8
  20. 2025Agentic Artificial Intelligence for Multistage Physics Experiments at a Large-Scale User Facility Particle Acceleratorcited by 6
  21. 2025Open WebUI: An Open, Extensible, and Usable Interface for AI Interactioncited by 6
  22. 2024Towards Unlocking Insights from Logbooks Using AIcited by 5
  23. 2024Towards Agentic AI on Particle Acceleratorscited by 5
  24. 2023Robust Errant Beam Prognostics with Conditional Modeling for Particle Acceleratorscited by 4
  25. 2023Resilient VAE: Unsupervised Anomaly Detection at the SLAC Linac Coherent Light Sourcecited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026VecTree-RAG: An Agentic Retrieval-Augmented Generation Framework Combining Vector and Tree Retrieval for Efficiency and Accuracy
  2. 2025SQuAI: Scientific Question-Answering with Multi-Agent Retrieval-Augmented Generation
  3. 2026An Agent-Oriented Pluggable Experience-RAG Skill for Experience-Driven Retrieval Strategy Orchestration
  4. 2026Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model
  5. 2026Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval
  6. 2026When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering
  7. 2026Agent-Enhanced Heterogeneous Graph RAG for Academic Question Answering
  8. 2026Re-ranking and Late Interaction Drive Retrieval Quality: A Controlled Comparison of RAG Strategies for Scientific Question Answering