mlsanityData to 14 Oct 2026

Evidence Interfaces Shape How Retrieval-Augmented Readers Use Support

arXiv 2607.17108 ↗First posted 19 Jul 2026cs.AIcited by 1

On the map: Retrieval-augmented generation → RAG robustness and adaptive retrieval

Abstract (arXiv)

In multi-hop RAG evaluation, a top-k answer score can hide two different failures: the retrieval window may drop part of the support chain, or it may contain support in a form the adapted reader does not use well. We call this reader-facing form of retrieved evidence an evidence interface. Using three support-annotated multi-hop QA benchmarks, we compare matched adapted readers trained with raw context, retrieval windows, and gold-support diagnostic renderings. These comparisons distinguish support-availability failures from remaining reader-interface effects. Top-k windows become interpretable only after checking whether the complete annotated support chain survives: when it does, short ranked windows can match or improve over raw context; when it does not, missing support explains much of the loss. Gold support-first improves matched readers; on 2Wiki and MuSiQue, a support-supervised ranker raises coverage and recovers raw-context quality at lower prompt cost, while retaining gold headroom. Support-removal checks further show that the gains rely on exposed evidence, not only answer priors. On support-annotated evaluations, top-k answer scores should therefore be reported together with complete-support coverage.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation. Builds on 13 of its 15 references that are in the map. First posted on arXiv 19 Jul 2026.

Builds on

13 of 15 references.

References · 13

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,236
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  5. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  6. 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,417
  7. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,282
  8. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 952
  9. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 916
  10. 2023REPLUG: Retrieval-Augmented Black-Box Language Modelscited by 792
  11. 2017Reading Wikipedia to Answer Open-Domain Questionscited by 777
  12. 2019Passage Re-ranking with BERTcited by 645
1 more
  1. 2024RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMscited by 202

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026What Would Fix This RAG Failure? Auditing Counterfactual Response with Paired Evidence Interventionscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Assessing the Downstream Utility of Evidence-Aware Retrieval in RAG
  2. 2026CUE-R: Beyond the Final Answer in Retrieval-Augmented Generation
  3. 2026Facet-Level Tracing of Evidence Uncertainty and Hallucination in RAG
  4. 2026ProvenAI: Provenance-Native Traces of Evidence in Generated Answers
  5. 2026Why Retrieval-Augmented Generation Fails: A Graph Perspective
  6. 2026PAGE-RAG: Provenance-Aware Graph Evidence Promotion for Fixed-Budget Multi-hop Retrieval-Augmented Generation
  7. 2026Recall Is Not Enough: A Reader-Context Diagnostic for Budget-Constrained Retrieval-Augmented Generation
  8. 2026RAG-Stress: Probing the Limits of Evidence Reliance in Retrieval-Augmented Generation