mlsanityData to 7 Oct 2026

Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretraining

arXiv 2510.22931 ↗First posted 27 Oct 2025cs.LG cs.AIcited by 2

On the map: Synthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping

Abstract (arXiv)

Continual Learning (CL) is essential for enabling self-evolving large language models (LLMs) to adapt and remain effective amid rapid knowledge growth. Yet, despite its importance, little attention has been given to establishing statistical reliability guarantees for LLMs under CL, particularly in the setting of continual domain pretraining (CDP). Conformal Prediction (CP) has shown promise in offering correctness guarantees for LLMs, but it faces major challenges in CDP: testing data often stems from unknown or shifting domain distributions, under which CP may no longer provide valid guarantees. Moreover, when high coverage is required, CP can yield excessively large prediction sets for unanswerable queries, reducing informativeness. To address these challenges, we introduce an adaptive rejection and non-exchangeable CP framework. Our method first estimates the distribution of questions across domains in the test set using transformer-based clustering, then reweights or resamples the calibration data accordingly. Building on this, adaptive rejection CP allows the LLM to selectively abstain from answering when its confidence or competence shifts significantly. Extensive experiments demonstrate that our framework enhances both the effectiveness and reliability of CP under CDP scenarios. Our code is available at: https://anonymous.4open.science/r/CPCL-8C12/

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Synthetic reasoning data and bootstrapping, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 23 of its 41 references that are in the map. First posted on arXiv 27 Oct 2025.

Builds on

23 of 41 references.

References · 23

  1. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  2. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  3. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  4. 2020Don't Stop Pretraining: Adapt Language Models to Domains and Taskscited by 1,548
  5. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 777
  6. 2020Aligning AI With Shared Human Valuescited by 706
  7. 2018An Empirical Study of Example Forgetting during Deep Neural Network Learningcited by 669
  8. 2019Conformal Prediction Under Covariate Shiftcited by 511
  9. 2024Continual Learning of Large Language Models: A Comprehensive Surveycited by 228
  10. 2024SaulLM-7B: A pioneering Large Language Model for Lawcited by 136
  11. 2023Conformal Prediction with Large Language Models for Multi-Choice Question Answeringcited by 131
  12. 2021DEMix Layers: Disentangling Domains for Modular Language Modelingcited by 130
11 more
  1. 2024Language Models with Conformal Factuality Guaranteescited by 128
  2. 2022ELLE: Efficient Lifelong Pre-training for Emerging Datacited by 70
  3. 2024API Is Enough: Conformal Prediction for Large Language Models Without Logit-Accesscited by 68
  4. 2024Conformal Prediction for Natural Language Processing: A Surveycited by 52
  5. 2020ECONET: Effective Continual Pretraining of Language Models for Event Temporal Reasoningcited by 38
  6. 2025Analyzing Uncertainty of LLM-as-a-Judge: Interval Evaluations with Conformal Predictioncited by 26
  7. 2023TRAQ: Trustworthy Retrieval Augmented Question Answering via Conformal Predictioncited by 23
  8. 2023Recyclable Tuning for Continual Pre-trainingcited by 17
  9. 2025Conformal Language Model Reasoning with Coherent Factualitycited by 14
  10. 2024Transformer-based Causal Language Models Perform Clusteringcited by 5
  11. 2022AF Adapter: Continual Pretraining for Building Chinese Biomedical Language Modelcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarificationcited by 12
  2. 2026SELAUR: Self Evolving LLM Agent via Uncertainty-aware Rewardscited by 9

Similar, not cited

Close in wording, no citation either way

  1. 2024ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees
  2. 2025CGES: Confidence-Guided Early Stopping for Efficient and Accurate Self-Consistency
  3. 2026Beyond Surface Statistics: Robust Conformal Prediction for LLMs via Internal Representations
  4. 2025Trusted Uncertainty in Large Language Models: A Unified Framework for Confidence Calibration and Risk-Controlled Refusal
  5. 2024Continual Learning for Large Language Models: A Survey
  6. 2025Beyond Exponential Decay: Rethinking Error Accumulation in Large Language Models
  7. 2026Time-Incremental Continued Pretraining of LLMs: Knowledge Updates Without Catastrophic Forgetting
  8. 2025Certified Self-Consistency: Statistical Guarantees and Test-Time Training for Reliable Reasoning in LLMs