mlsanityData to 7 Oct 2026

With Argus Eyes: Assessing Retrieval Gaps via Uncertainty Scoring to Detect and Remedy Retrieval Blind Spots

arXiv 2602.09616 ↗First posted 10 Feb 2026cs.IR cs.AIcited by 1

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

Abstract (arXiv)

Reliable retrieval-augmented generation (RAG) systems depend fundamentally on the retriever's ability to find relevant information. We show that neural retrievers used in RAG systems have blind spots, which we define as the failure to retrieve entities that are relevant to the query, but have low similarity to the query embedding. We investigate the training-induced biases that cause such blind spot entities to be mapped to inaccessible parts of the embedding space, resulting in low retrievability. Using a large-scale dataset constructed from Wikidata relations and first paragraphs of Wikipedia, and our proposed Retrieval Probability Score (RPS), we show that blind spot risk in standard retrievers (e.g., CONTRIEVER, REASONIR) can be predicted pre-index from entity embedding geometry, avoiding expensive retrieval evaluations. To address these blind spots, we introduce ARGUS, a pipeline that enables the retrievability of high-risk (low-RPS) entities through targeted document augmentation from a knowledge base (KB), first paragraphs of Wikipedia, in our case. Extensive experiments on BRIGHT, IMPLIRET, and RAR-B show that ARGUS achieves consistent improvements across all evaluated retrievers (averaging +3.4 nDCG@5 and +4.5 nDCG@10 absolute points), with substantially larger gains in challenging subsets. These results establish that preemptively remedying blind spots is critical for building robust and trustworthy RAG systems.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation. Builds on 58 of its 72 references that are in the map. First posted on arXiv 10 Feb 2026.

Builds on

58 of 72 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Trustworthiness in Retrieval-Augmented Generation Systems: A Survey16 Sep
  2. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection17 Oct

Other references · 56

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  3. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  4. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
  5. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  6. 2022Survey of Hallucination in Natural Language Generationcited by 2,056
  7. 2016MS MARCO: A Human Generated MAchine Reading COmprehension Datasetcited by 1,591
  8. 2023Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injectioncited by 1,339
  9. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,225
  10. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  11. 2025Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Modelscited by 1,175
  12. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
38 more
  1. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,050
  2. 2020Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrievalcited by 846
  3. 2019How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddingscited by 770
  4. 2021Retrieval Augmentation Reduces Hallucination in Conversationcited by 691
  5. 2019Passage Re-ranking with BERTcited by 629
  6. 2022MTEB: Massive Text Embedding Benchmarkcited by 623
  7. 2022Precise Zero-Shot Dense Retrieval without Relevance Labelscited by 508
  8. 2024NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Modelscited by 466
  9. 2021ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interactioncited by 464
  10. 2022Large Language Models Struggle to Learn Long-Tail Knowledgecited by 452
  11. 2023Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflictscited by 333
  12. 2024PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Modelscited by 251
  13. 2019Representation Degeneration Problem in Training Natural Language Generation Modelscited by 250
  14. 2024Generative Representational Instruction Tuningcited by 227
  15. 2019Document Expansion by Query Predictioncited by 225
  16. 2023Query Rewriting for Retrieval-Augmented Large Language Modelscited by 211
  17. 2023Poisoning Retrieval Corpora by Injecting Adversarial Passagescited by 153
  18. 2024BRIGHT: A Realistic and Challenging Benchmark for Reasoning-Intensive Retrievalcited by 153
  19. 2023Query2doc: Query Expansion with Large Language Modelscited by 146
  20. 2024Certifiably Robust RAG against Retrieval Corruptioncited by 144
  21. 2021Simple Entity-Centric Questions Challenge Dense Retrieverscited by 140
  22. 2024jina-embeddings-v3: Multilingual Embeddings With Task LoRAcited by 138
  23. 2025NoLiMa: Long-Context Evaluation Beyond Literal Matchingcited by 120
  24. 2025On the Theoretical Limitations of Embedding-Based Retrievalcited by 98
  25. 2024BM25S: Orders of magnitude faster lexical search via eager sparse scoringcited by 87
  26. 2025ReasonIR: Training Retrievers for Reasoning Taskscited by 76
  27. 2020The Curse of Dense Low-Dimensional Information Retrieval for Large Index Sizescited by 48
  28. 2021Societal Biases in Retrieved Contents: Measurement Framework and Adversarial Mitigation for BERT Rankerscited by 34
  29. 2024RAR-b: Reasoning as Retrieval Benchmarkcited by 33
  30. 2025Towards Trustworthy Retrieval Augmented Generation for Large Language Models: A Surveycited by 29
  31. 2020Do Neural Ranking Models Intensify Gender Bias?cited by 28
  32. 2023Doc2Query--: When Less is Morecited by 20
  33. 2022COCO-DR: Combating Distribution Shifts in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learningcited by 19
  34. 2024Robust Neural Information Retrieval: An Adversarial and Out-of-distribution Perspectivecited by 15
  35. 2025ReasonEmbed: Enhanced Text Embeddings for Reasoning-Intensive Document Retrievalcited by 15
  36. 2024Systematic Evaluation of Neural Retrieval Models on the Touch'e 2020 Argument Retrieval Subset of BEIRcited by 9
  37. 2024Universal Novelty Detection Through Adaptive Contrastive Learningcited by 9
  38. 2025RODEO: Robust Outlier Detection via Exposing Adaptive Out-of-Distribution Samplescited by 7

50 most cited of 56 All 56 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Better Nearest Neighbor Graph Indices via (Efficient) LLM-Guided Pruningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Collapse of Dense Retrievers: Short, Early, and Literal Biases Outranking Factual Evidence
  2. 2025RADIANT: Retrieval AugmenteD entIty-context AligNmenT -- Introducing RAG-ability and Entity-Context Divergence
  3. 2025Retrieval-Augmented Generation: A Comprehensive Survey of Architectures, Enhancements, and Robustness Frontiers
  4. 2026Revisiting RAG Retrievers: An Information Theoretic Benchmark
  5. 2024Beyond Benchmarks: Evaluating Embedding Model Similarity for Retrieval Augmented Generation Systems
  6. 2026Beyond the Reranker: Do RAG Retrieval Enhancements Help Once a Strong Reranker Is Present?
  7. 2026Uncertainty-Aware Hybrid Retrieval for Long-Document RAG
  8. 2024Astute RAG: Overcoming Imperfect Retrieval Augmentation and Knowledge Conflicts for Large Language Models