mlsanityData to 7 Oct 2026

BalanceRAG: Joint Risk Calibration for Cascaded Retrieval-Augmented Generation

arXiv 2605.20084 ↗First posted 19 May 2026cs.CL cs.AIcited by 2

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

Abstract (arXiv)

Large language models (LLMs) can enhance factuality via retrieval-augmented generation (RAG), but applying RAG to every query is unnecessary when the model-only answer is reliable. This motivates cascaded RAG: each query is first handled by an LLM-only branch, escalated to a RAG fallback only if the primary branch is uncertain, and abstained from when neither branch is sufficiently trustworthy. However, calibrating such cascades stage by stage may be conservative, since the final utility depends on joint uncertainty thresholding of LLM-only and RAG. In this work, we develop BalanceRAG to certify threshold pairs at a target risk level. Given uncertainty scores from the two branches, BalanceRAG frames each threshold pair as an operating point on a two-dimensional lattice and identifies safe operating points using sequential graphical testing. This enables risk-adaptive threshold calibration, controlling the system-level error rate among accepted points, while retaining more examples. Furthermore, BalanceRAG extends to multi-risk calibration, allowing retrieval usage to be bounded together with the selection-conditioned risk. Experiments on three open-domain question answering (QA) benchmarks across multiple LLM backbones demonstrate that BalanceRAG meets prescribed risk levels, preserves higher coverage and more accepted correct examples, and reduces unnecessary retrieval calls compared with always-on RAG.

Key facts. Cited by 2 papers 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 20 of its 41 references that are in the map. First posted on arXiv 19 May 2026.

Builds on

20 of 41 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity21 Mar
  2. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection17 Oct

Other references · 18

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  4. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  5. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  6. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,538
  7. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
  8. 2024Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approachcited by 117
  9. 2024Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreementcited by 99
  10. 2024VL-Uncertainty: Detecting Hallucination in Large Vision-Language Model via Uncertainty Estimationcited by 71
  11. 2024ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guaranteescited by 62
  12. 2024Conformal Alignment: Knowing When to Trust Foundation Models with Guaranteescited by 50
6 more
  1. 2025MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Modelscited by 33
  2. 2024MBA-RAG: a Bandit Approach for Adaptive Retrieval-Augmented Generation through Question Complexitycited by 28
  3. 2025COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guaranteescited by 20
  4. 2026Set-Valued Prediction for Large Language Models with Feasibility-Aware Coverage Guaranteescited by 3
  5. 2025Coverage-Guaranteed Speech Emotion Recognition via Calibrated Uncertainty-Adaptive Prediction Setscited by 2
  6. 2025SkewRoute: Training-Free LLM Routing for Knowledge Graph Retrieval-Augmented Generation via Score Skewness of Retrieved Contextcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Uncertainty-Aware Abstention in Large Language Models with Provable Alignment Guaranteescited by 1
  2. 2026Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Studycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026URAG: A Benchmark for Uncertainty Quantification in Retrieval-Augmented Large Language Models
  2. 2025Investigating the Robustness of Retrieval-Augmented Generation at the Query Level
  3. 2025Retrieval-Augmented Generation: A Comprehensive Survey of Architectures, Enhancements, and Robustness Frontiers
  4. 2025QuCo-RAG: Quantifying Uncertainty from the Pre-training Corpus for Dynamic Retrieval-Augmented Generation
  5. 2025SGIC: A Self-Guided Iterative Calibration Framework for RAG
  6. 2025Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home
  7. 2025Response Quality Assessment for Retrieval-Augmented Generation via Conditional Conformal Factuality
  8. 2026CQC-RAG: Robust Retrieval-Augmented Generation via Cross-Query Consistency