mlsanityData to 7 Oct 2026

Beyond the Query: Do Retrieval Signals Improve Adaptive Multimodal RAG Routing?

arXiv 2609.12437 ↗First posted 11 Sep 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Adaptive RAG often uses retrieval-time signals to decide whether another retrieval, reranking, or multimodal step should run. We ask whether these signals add routing value once the query itself is already known. Across document, audio, and video RAG, we compare matched query-only and query+retrieval routers while holding the optional actions, router family, training procedure, and evaluation fixed. On the held-out final evaluation, adding the tested retrieval signals does not produce a reliable routing improvement over the query-only baseline. Some retrieval signals are associated with whether a later step will help, but that predictability does not consistently lead to bet- ter RUN/SKIP decisions. The main lesson is therefore methodological: retrieval-state features should not be credited with routing value unless they improve over a matched query-only control. Our results do not show that routing or retrieval state is generally useless; they show that the incremental value of retrieval signals must be demonstrated rather than assumed.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 18 of its 26 references that are in the map. First posted on arXiv 11 Sep 2026.

Builds on

18 of 26 references.

References · 18

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,800
  3. 2020DocVQA: A Dataset for VQA on Document Imagescited by 1,288
  4. 2021NExT-QA:Next Phase of Question-Answering to Explaining Temporal Actionscited by 735
  5. 2025Qwen2.5-Omni Technical Reportcited by 704
  6. 2023Active Retrieval Augmented Generationcited by 506
  7. 2024ColPali: Efficient Document Retrieval with Vision Language Modelscited by 190
  8. 2022Clotho-AQA: A Crowdsourced Dataset for Audio Question Answeringcited by 84
  9. 2024DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Modelscited by 63
  10. 2024MBA-RAG: a Bandit Approach for Adaptive Retrieval-Augmented Generation through Question Complexitycited by 28
  11. 2025DioR: Adaptive Cognitive Detection and Contextual Retrieval Optimization for Dynamic Retrieval-Augmented Generationcited by 7
  12. 2025LAD-RAG: Layout-aware Dynamic RAG for Visually-Rich Document Understandingcited by 7
6 more
  1. 2025RouteRAG: Efficient Retrieval-Augmented Generation from Text and Graph via Reinforcement Learningcited by 5
  2. 2023DF2: Distribution-Free Decision-Focused Learningcited by 2
  3. 2026When Should Active RAG Retrieve? A Budget-Aware Evaluation of Utility, Calibration, and Costcited by 2
  4. 2026Adaptive Re-Rankingcited by 1
  5. 2026Modality Relevance is not Modality Utility: Post-hoc Selective Modality Escalation for Cost-Aware Multimodal RAGcited by 1
  6. 2026SAGE: SLO-Aware Adaptive Retrieval for Production RAG Systemscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Cost-Aware Query Routing in RAG: Empirical Analysis of Retrieval Depth Tradeoffs
  2. 2026How You Ask Matters! Adaptive RAG Robustness to Query Variations
  3. 2026RAGRouter-Bench: A Dataset and Benchmark for Adaptive RAG Routing
  4. 2026When Should Multi-Round RAG Stop? Structured Stopping Judgments and Retrieval Reduction in Search-R1
  5. 2026Rethinking RAG in Long Videos: What to Retrieve and How to Use It?
  6. 2026Feedback Adaptation for Retrieval-Augmented Generation
  7. 2025Test-time Corpus Feedback: From Retrieval to RAG
  8. 2024Self-adaptive Multimodal Retrieval-Augmented Generation