cs.LGJun 16, 2025

Online Conformal Abstention Under Adversarial Bandit Feedback

Authors: Minjae Lee, Yoonjae Jung, Sangdon Park

Organizations: GSAI POSTECH · GSAI & CSE POSTECH

Abstract

As interactive generative systems are increasingly deployed in real-world applications, their tendency to generate unreliable or false responses raises serious concerns. Conformal abstention mitigates this risk by ensuring that the system answers only when confident. However, real-world deployments typically provide only partial user feedback (e.g., thumbs up/down) on the selected response and often operate in non-stationary or adversarial environments, for which effective learning methods are largely missing. To bridge this gap, we propose ExAUL, a novel online learning framework for conformal abstention with adversarial and partial feedback. Technically, we introduce (i) a novel conversion lemma that translates the regret of any bandit algorithm into a selective risk (SR) bound, and (ii) feedback unlocking, a strategy that exploits the structure of conformal abstention to extract additional learning signals from partial feedback. We prove that ExAUL achieves a regret bound of O(Tln⁡∣H∣)O(\sqrt{T \ln |H|}), which translates into an O(T)O(\sqrt{T}) bound on SR violation control, matching the controllability of full-information settings despite receiving only partial feedback. While applicable to general generative tasks, we demonstrate the efficacy of ExAUL for ensuring the reliability of Large Language Models (LLMs) through empirical validation on question-answering tasks across diverse non-stationary and adversarial settings. Our results demonstrate that ExAUL robustly controls the SR while maintaining competitive answering coverage. Code and datasets are provided at https://github.com/trustml-lab/oca.

Figures & tables

Appendix figures & tables37 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Geometry-Calibrated Conformal Abstention for Language Models

    Apr 30, 2026Rui Xu, Yi Chen, Sihong Xie +1Uncertainty QuantificationSelective Prediction

  2. Conf-Gen: Conformal Uncertainty Quantification for Generative Models

    May 27, 2026Gabriel Loaiza-Ganem, Kevin Zhang, Wei Cui +2Uncertainty QuantificationConformal Prediction

  3. Strong yet Controllable LLM Abstention without Learning to Abstain

    Oct 12, 2026Junwei Zhang, Yifei Wang, Wanli Yang +6Factual QALLM Abstention