Semantic Self-Distillation for Language Model Uncertainty

• URL: http://arxiv.org/abs/2602.04577v1
• Date: Wed, 04 Feb 2026 14:03:28 GMT
• Title: Semantic Self-Distillation for Language Model Uncertainty
• Authors: Edward Phillips, Sean Wu, Boyan Gao, David A. Clifton,
• Abstract summary: We show that lightweight student models can estimate a prompt-conditioned uncertainty before a language model generates an answer token.<n>The entropy of this distribution provides an effective uncertainty signal for hallucination prediction and the probability density allows candidate answers to be evaluated for reliability.<n>On TriviaQA, our student models match or outperform finite-sample semantic dispersion for hallucination prediction and provide a strong signal for out-of-domain answer detection.
• Score: 19.97226069762587
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models present challenges for principled uncertainty quantification, in part due to their complexity and the diversity of their outputs. Semantic dispersion, or the variance in the meaning of sampled answers, has been proposed as a useful proxy for model uncertainty, but the associated computational cost prohibits its use in latency-critical applications. We show that sampled semantic distributions can be distilled into lightweight student models which estimate a prompt-conditioned uncertainty before the language model generates an answer token. The student model predicts a semantic distribution over possible answers; the entropy of this distribution provides an effective uncertainty signal for hallucination prediction, and the probability density allows candidate answers to be evaluated for reliability. On TriviaQA, our student models match or outperform finite-sample semantic dispersion for hallucination prediction and provide a strong signal for out-of-domain answer detection. We term this technique Semantic Self-Distillation (SSD), which we suggest provides a general framework for distilling predictive uncertainty in complex output spaces beyond language.

Related papers

• The Illusion of Certainty: Uncertainty quantification for LLMs fails under ambiguity [48.899855816199484]
  We introduce MAQA* and AmbigQA*, the first ambiguous question-answering (QA) datasets equipped with ground-truth answer distributions.<n>We show that predictive-distribution and ensemble-based estimators are fundamentally limited under ambiguity.
  arXiv  Detail & Related papers  (2025-11-06T14:46:35Z)
• Efficient semantic uncertainty quantification in language models via diversity-steered sampling [46.23327887393273]
  We introduce a diversity-steered sampler that discourages semantically redundant outputs during decoding.<n>Key idea is to inject a continuous semantic-similarity penalty into the model's proposal distribution.<n>Being modular and requiring no gradient access to the base LLM, the framework promises to serve as a drop-in enhancement for uncertainty estimation.
  arXiv  Detail & Related papers  (2025-10-24T10:06:21Z)
• Variability Need Not Imply Error: The Case of Adequate but Semantically Distinct Responses [7.581259361859477]
  Uncertainty quantification tools can be used to reject a response when the model is uncertain'<n>We estimate the Probability the model assigns to Adequate Responses (PROBAR)<n>We find PROBAR to outperform semantic entropy across prompts with varying degrees of ambiguity/open-endedness.
  arXiv  Detail & Related papers  (2024-12-20T09:02:26Z)
• Decomposing Uncertainty for Large Language Models through Input Clarification Ensembling [69.83976050879318]
  In large language models (LLMs), identifying sources of uncertainty is an important step toward improving reliability, trustworthiness, and interpretability. In this paper, we introduce an uncertainty decomposition framework for LLMs, called input clarification ensembling. Our approach generates a set of clarifications for the input, feeds them into an LLM, and ensembles the corresponding predictions.
  arXiv  Detail & Related papers  (2023-11-15T05:58:35Z)
• Quantification of Predictive Uncertainty via Inference-Time Sampling [57.749601811982096]
  We propose a post-hoc sampling strategy for estimating predictive uncertainty accounting for data ambiguity. The method can generate different plausible outputs for a given input and does not assume parametric forms of predictive distributions.
  arXiv  Detail & Related papers  (2023-08-03T12:43:21Z)
• User-defined Event Sampling and Uncertainty Quantification in Diffusion Models for Physical Dynamical Systems [49.75149094527068]
  We show that diffusion models can be adapted to make predictions and provide uncertainty quantification for chaotic dynamical systems. We develop a probabilistic approximation scheme for the conditional score function which converges to the true distribution as the noise level decreases. We are able to sample conditionally on nonlinear userdefined events at inference time, and matches data statistics even when sampling from the tails of the distribution.
  arXiv  Detail & Related papers  (2023-06-13T03:42:03Z)
• Uncertainty Estimation in Deep Speech Enhancement Using Complex Gaussian Mixture Models [19.442685015494316]
  Single-channel deep speech enhancement approaches often estimate a single multiplicative mask to extract clean speech without a measure of its accuracy. We propose to quantify the uncertainty associated with clean speech estimates in neural network-based speech enhancement.
  arXiv  Detail & Related papers  (2022-12-09T13:03:09Z)
• The Hidden Uncertainty in a Neural Networks Activations [105.4223982696279]
  The distribution of a neural network's latent representations has been successfully used to detect out-of-distribution (OOD) data. This work investigates whether this distribution correlates with a model's epistemic uncertainty, thus indicating its ability to generalise to novel inputs.
  arXiv  Detail & Related papers  (2020-12-05T17:30:35Z)
• Modal Uncertainty Estimation via Discrete Latent Representation [4.246061945756033]
  We introduce a deep learning framework that learns the one-to-many mappings between the inputs and outputs, together with faithful uncertainty measures. Our framework demonstrates significantly more accurate uncertainty estimation than the current state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-07-25T05:29:34Z)

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.