mlsanityData to 14 Oct 2026

NOVA: NOise-aware Verbal Confidence CAlibration for Robust Large Language Models in RAG Systems

arXiv 2601.11004 ↗First posted 16 Jan 2026cs.CLcited by 6

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

Abstract (arXiv)

Accurately assessing model confidence is essential for deploying large language models (LLMs) in mission-critical factual domains. While retrieval-augmented generation (RAG) is widely adopted to improve grounding, confidence calibration in RAG settings remains poorly understood. We conduct a systematic study across four benchmarks, revealing that LLMs exhibit poor calibration performance especially when noisy contexts are retrieved. Specifically, contradictory or irrelevant evidence tends to exacerbate the model's overconfidence issue. To address this, we propose NOVA Rules (NOise-Aware Verbal Confidence CAlibration Rules) to provide a principled foundation for resolving overconfidence under noise. We further design NOVA, a noise-aware calibration framework that synthesizes supervision from ~2K HotpotQA examples guided by these rules. By performing supervised fine-tuning (SFT) with this data, NOVA equips models with intrinsic noise awareness without relying on stronger teacher models. Empirical results show that NOVA yields substantial gains, improving ECE scores by 10.9% in-domain and 8.0% out-of-domain. By bridging the gap between retrieval noise and verbal calibration, NOVA paves the way for both accurate and epistemically reliable LLMs.

Key facts. Cited by 6 papers 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 64 of its 118 references that are in the map. First posted on arXiv 16 Jan 2026.

Builds on

64 of 118 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025TruthfulRAG: Resolving Factual-level Conflicts in Retrieval-Augmented Generation with Knowledge Graphs13 Nov
  2. 2025Uncertainty Quantification for Retrieval-Augmented Reasoning13 Oct
  3. 2025SGIC: A Self-Guided Iterative Calibration Framework for RAG19 Jun
  4. 2025Divide-Then-Align: Honest Alignment based on the Knowledge Boundary of RAG27 May
  5. 2025LLM-based Corroborating and Refuting Evidence Retrieval for Scientific Claim Verification11 Mar
  6. 2025Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home22 Jan
  7. 2024RAG-RewardBench: Benchmarking Reward Models in Retrieval Augmented Generation for Preference Alignment18 Dec
  8. 2024Quantifying reliance on external information over parametric knowledge during Retrieval Augmented Generation (RAG) using mechanistic analysis1 Oct
  9. 2024WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia19 Jun
  10. 2024From RAGs to rich parameters: Probing how language models utilize external knowledge over parametric information for factual queries18 Jun
  11. 2024Enhancing Noise Robustness of Retrieval-Augmented Language Models with Adaptive Adversarial Training31 May
  12. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection17 Oct

Other references · 52

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  4. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  5. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  6. 2017On Calibration of Modern Neural Networkscited by 4,960
  7. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  8. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  9. 2020Learning to summarize from human feedbackcited by 2,588
  10. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  11. 2024LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Modelscited by 1,464
  12. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,287
38 more
  1. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,216
  2. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  3. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  4. 2022Teaching Models to Express Their Uncertainty in Wordscited by 751
  5. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 743
  6. 2025BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agentscited by 647
  7. 2023Making Retrieval-Augmented Language Models Robust to Irrelevant Contextcited by 369
  8. 2024Knowledge Conflicts for LLMs: A Surveycited by 283
  9. 2024The Power of Noise: Redefining Retrieval for RAG Systemscited by 281
  10. 2021AI Chains: Transparent and Controllable Human-AI Interaction by Chaining Large Language Model Promptscited by 268
  11. 2023Do Large Language Models Know What They Don't Know?cited by 243
  12. 2020Unsupervised Quality Estimation for Neural Machine Translationcited by 211
  13. 2023Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form Large Language Modelscited by 150
  14. 2023LM-Polygraph: Uncertainty Estimation for Language Modelscited by 131
  15. 2023Quantifying Uncertainty in Answers from any Language Model and Enhancing their Trustworthinesscited by 122
  16. 2025Beyond Binary Rewards: Training LMs to Reason About Their Uncertaintycited by 117
  17. 2024SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationalescited by 105
  18. 2025Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversitycited by 105
  19. 2025HalluLens: LLM Hallucination Benchmarkcited by 92
  20. 2024On Verbalized Confidence Scores for LLMscited by 83
  21. 2019Calibration tests in multi-class classification: A unifying frameworkcited by 79
  22. 2025Efficient Test-Time Scaling via Self-Calibrationcited by 75
  23. 2024Calibrating Large Language Models with Sample Consistencycited by 73
  24. 2019Measuring the Quality of Explanations: The System Causability Scale (SCS). Comparing Human and Machine Explanationscited by 64
  25. 2025A Survey of Uncertainty Estimation Methods on Large Language Modelscited by 62
  26. 2024Calibrating Language Models with Adaptive Temperature Scalingcited by 58
  27. 2025ConfTuner: Training Large Language Models to Express Their Confidence Verballycited by 39
  28. 2025CostBench: Evaluating Multi-Turn Cost-Optimal Planning and Adaptation in Dynamic Environments for LLM Tool-Use Agentscited by 33
  29. 2024RL-GPT: Integrating Reinforcement Learning and Code-as-policycited by 32
  30. 2025UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarioscited by 26
  31. 2025GaRAGe: A Benchmark with Grounding Annotations for RAG Evaluationcited by 18
  32. 2025Revisiting Epistemic Markers in Confidence Estimation: Can Markers Accurately Reflect Large Language Models' Uncertainty?cited by 16
  33. 2024Towards Understanding Systems Trade-offs in Retrieval-Augmented Generation Model Inferencecited by 15
  34. 2025Stream RAG: Instant and Accurate Spoken Dialogue Systems with Streaming Tool Usagecited by 15
  35. 2025Why Uncertainty Estimation Methods Fall Short in RAG: An Axiomatic Analysiscited by 14
  36. 2024ComparisonQA: Evaluating Factuality Robustness of LLMs Through Knowledge Frequency Control and Uncertaintycited by 10
  37. 2022A Benchmark for Generalizable and Interpretable Temporal Question Answering over Knowledge Basescited by 9
  38. 2024Understanding the Impact of Confidence in Retrieval Augmented Generation: A Case Study in the Medical Domaincited by 9

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystemscited by 11
  2. 2026AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraintscited by 7
  3. 2026CreativityBench: Evaluating Agent Creative Reasoning via Affordance-Based Tool Repurposingcited by 4
  4. 2026When Confidence Takes the Wrong Path: Diagnosing Retrieval-State Lock-In in RAGcited by 2
  5. 2026EvolvingWorld: An Open-Schema Framework for Co-Evolving Role-Play Agents and World Model in Interactive Literary Worldcited by 1
  6. 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. 2026Confident in a Confidence Score: Investigating the Sensitivity of Confidence Scores to Supervised Fine-Tuning
  2. 2026ConfidenceBench: Evaluating Confidence Calibration in Large Language Models
  3. 2024Calibrating the Confidence of Large Language Models by Eliciting Fidelity
  4. 2025Trusted Uncertainty in Large Language Models: A Unified Framework for Confidence Calibration and Risk-Controlled Refusal
  5. 2023On the Calibration of Large Language Models and Alignment
  6. 2024Calibrating Long-form Generations from Large Language Models
  7. 2025Direct Confidence Alignment: Aligning Verbalized Confidence with Internal Confidence In Large Language Models
  8. 2025REHEARSE: Experiential Rehearsal for Verbal Confidence Calibration in Large Language Models