Efficient Epistemic Uncertainty Estimation for Large Language Models via Knowledge Distillation

• URL: http://arxiv.org/abs/2602.01956v1
• Date: Mon, 02 Feb 2026 11:03:37 GMT
• Title: Efficient Epistemic Uncertainty Estimation for Large Language Models via Knowledge Distillation
• Authors: Seonghyeon Park, Jewon Yeom, Jaewon Sok, Jeongjae Park, Heejun Kim, Taesup Kim,
• Abstract summary: Epistemic Uncertainty(EU) via Deep Ensembles is computationally prohibitive at the scale of modern models.<n>We propose a framework that leverages the small draft models to efficiently estimate token-level EU, bypassing the need for full-scale ensembling.
• Score: 8.883246421566243
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Quantifying uncertainty in Large Language Models (LLMs) is essential for mitigating hallucinations and enabling risk-aware deployment in safety-critical tasks. However, estimating Epistemic Uncertainty(EU) via Deep Ensembles is computationally prohibitive at the scale of modern models. We propose a framework that leverages the small draft models to efficiently estimate token-level EU, bypassing the need for full-scale ensembling. Theoretically grounded in a Bias-Variance Decomposition, our approach approximates EU via Jensen-Shannon divergence among drafts (variance proxy) and KL divergence between the draft mixture and the target (bias proxy). To further ensure accuracy without significant overhead, we introduce Online Stochastic Distillation (OSD) to efficiently approximate target aggregation and the Data-Diverse Drafts (DDD) strategy to enhance draft diversity for better target approximation. Extensive experiments on GSM8K demonstrate that our method reduces the estimation error (RMSE) by up to 37% compared to baselines. Crucially, our approach achieves Hallucination Detection performance competitive with heavy perturbation-based methods like TokUR while incurring negligible inference costs, offering a practical solution for uncertainty-aware LLM deployment.

Related papers

• Exploring Semantic-constrained Adversarial Example with Instruction Uncertainty Reduction [51.50282796099369]
  This paper develops a multi-dimensional instruction uncertainty reduction framework to generate semantically constrained adversarial examples.<n>By predicting the language-guided sampling process, the optimization process will be stabilized by the designed ResAdv-DDIM sampler.<n>We realize the reference-free generation of semantically constrained 3D adversarial examples for the first time.
  arXiv  Detail & Related papers  (2025-10-27T04:02:52Z)
• ESI: Epistemic Uncertainty Quantification via Semantic-preserving Intervention for Large Language Models [23.44710972442814]
  Uncertainty Quantification (UQ) is a promising approach to improve model reliability, yet the uncertainty of Large Language Models (LLMs) is non-trivial.<n>We propose a novel grey-box uncertainty quantification method that measures the variation in model outputs before and after semantic-preserving intervention.
  arXiv  Detail & Related papers  (2025-10-15T02:46:43Z)
• Towards Reliable LLM-based Robot Planning via Combined Uncertainty Estimation [68.106428321492]
  Large language models (LLMs) demonstrate advanced reasoning abilities, enabling robots to understand natural language instructions and generate high-level plans with appropriate grounding.<n>LLMs hallucinations present a significant challenge, often leading to overconfident yet potentially misaligned or unsafe plans.<n>We present Combined Uncertainty estimation for Reliable Embodied planning (CURE), which decomposes the uncertainty into epistemic and intrinsic uncertainty, each estimated separately.
  arXiv  Detail & Related papers  (2025-10-09T10:26:58Z)
• Decoding Uncertainty: The Impact of Decoding Strategies for Uncertainty Estimation in Large Language Models [58.198220611190884]
  We investigate the impact of decoding strategies on uncertainty estimation in Large Language Models (LLMs)<n>Our experiments show that Contrastive Search, which mitigates repetition, yields better uncertainty estimates on average across a range of preference-aligned LLMs.
  arXiv  Detail & Related papers  (2025-09-20T13:48:13Z)
• Quantifying Adversarial Uncertainty in Evidential Deep Learning using Conflict Resolution [2.321323878201932]
  Conflict-aware Evidential Deep Learning (C-EDL) is a lightweight post-hoc uncertainty quantification approach.<n>C-EDL generates diverse, task-preserving transformations per input and quantifies disagreement to calibrate uncertainty estimates.<n>Our experimental evaluation shows that C-EDL significantly outperforms state-of-the-art EDL variants and competitive baselines.
  arXiv  Detail & Related papers  (2025-06-06T10:06:23Z)
• Towards Harmonized Uncertainty Estimation for Large Language Models [22.58034272573749]
  It is essential to quantify the reliability of their generations through uncertainty estimation.<n>We propose CUE (Corrector for Uncertainty Estimation): A straightforward yet effective method that employs a lightweight model trained on data aligned with the target LLM's performance to adjust uncertainty scores.
  arXiv  Detail & Related papers  (2025-05-25T10:17:57Z)
• Improving Uncertainty Quantification in Large Language Models via Semantic Embeddings [11.33157177182775]
  Accurately quantifying uncertainty in large language models (LLMs) is crucial for their reliable deployment. Current state-of-the-art methods for measuring semantic uncertainty in LLMs rely on strict bidirectional entailment criteria. We propose a novel approach that leverages semantic embeddings to achieve smoother and more robust estimation of semantic uncertainty.
  arXiv  Detail & Related papers  (2024-10-30T04:41:46Z)
• Revisiting Essential and Nonessential Settings of Evidential Deep Learning [70.82728812001807]
  Evidential Deep Learning (EDL) is an emerging method for uncertainty estimation. We propose Re-EDL, a simplified yet more effective variant of EDL.
  arXiv  Detail & Related papers  (2024-10-01T04:27:07Z)
• Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Models [104.55763564037831]
  We train a regression model that leverages attention maps, probabilities on the current generation step, and recurrently computed uncertainty scores from previously generated tokens.<n>Our evaluation shows that the proposed method is highly effective for selective generation, achieving substantial improvements over rivaling unsupervised and supervised approaches.
  arXiv  Detail & Related papers  (2024-08-20T09:42:26Z)
• Uncertainty Estimation by Fisher Information-based Evidential Deep Learning [61.94125052118442]
  Uncertainty estimation is a key factor that makes deep learning reliable in practical applications. We propose a novel method, Fisher Information-based Evidential Deep Learning ($mathcalI$-EDL) In particular, we introduce Fisher Information Matrix (FIM) to measure the informativeness of evidence carried by each sample, according to which we can dynamically reweight the objective loss terms to make the network more focused on the representation learning of uncertain classes.
  arXiv  Detail & Related papers  (2023-03-03T16:12:59Z)
• Distributionally Robust Model-Based Offline Reinforcement Learning with Near-Optimal Sample Complexity [39.886149789339335]
  offline reinforcement learning aims to learn to perform decision making from history data without active exploration. Due to uncertainties and variabilities of the environment, it is critical to learn a robust policy that performs well even when the deployed environment deviates from the nominal one used to collect the history dataset. We consider a distributionally robust formulation of offline RL, focusing on robust Markov decision processes with an uncertainty set specified by the Kullback-Leibler divergence in both finite-horizon and infinite-horizon settings.
  arXiv  Detail & Related papers  (2022-08-11T11:55:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.