Uncertainty-Aware Decoding with Minimum Bayes Risk

• URL: http://arxiv.org/abs/2503.05318v1
• Date: Fri, 07 Mar 2025 10:55:12 GMT
• Title: Uncertainty-Aware Decoding with Minimum Bayes Risk
• Authors: Nico Daheim, Clara Meister, Thomas Möllenhoff, Iryna Gurevych,
• Abstract summary: We show how Minimum Bayes Risk decoding, which selects model generations according to an expected risk, can be generalized into a principled uncertainty-aware decoding method.<n>We show that this modified expected risk is useful for both choosing outputs and deciding when to abstain from generation and can provide improvements without incurring overhead.
• Score: 70.6645260214115
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite their outstanding performance in the majority of scenarios, contemporary language models still occasionally generate undesirable outputs, for example, hallucinated text. While such behaviors have previously been linked to uncertainty, there is a notable lack of methods that actively consider uncertainty during text generation. In this work, we show how Minimum Bayes Risk (MBR) decoding, which selects model generations according to an expected risk, can be generalized into a principled uncertainty-aware decoding method. In short, we account for model uncertainty during decoding by incorporating a posterior over model parameters into MBR's computation of expected risk. We show that this modified expected risk is useful for both choosing outputs and deciding when to abstain from generation and can provide improvements without incurring overhead. We benchmark different methods for learning posteriors and show that performance improves with prediction diversity. We release our code publicly.

Related papers

• Distributionally Robust Optimisation with Bayesian Ambiguity Sets [8.642152250082368]
  We introduce Distributionally Robust optimisation with Bayesian Ambiguity Sets (DRO-BAS) DRO-BAS hedges against uncertainty in the model by optimising the worst-case risk over a posterior-informed ambiguity set. We show that our method admits a closed-form dual representation for many exponential family members.
  arXiv  Detail & Related papers  (2024-09-05T12:59:38Z)
• Beyond the Norms: Detecting Prediction Errors in Regression Models [26.178065248948773]
  This paper tackles the challenge of detecting unreliable behavior in regression algorithms. We introduce the notion of unreliability in regression, when the output of the regressor exceeds a specified discrepancy (or error) We show empirical improvements in error detection for multiple regression tasks, consistently outperforming popular baseline approaches.
  arXiv  Detail & Related papers  (2024-06-11T05:51:44Z)
• Cycles of Thought: Measuring LLM Confidence through Stable Explanations [53.15438489398938]
  Large language models (LLMs) can reach and even surpass human-level accuracy on a variety of benchmarks, but their overconfidence in incorrect responses is still a well-documented failure mode. We propose a framework for measuring an LLM's uncertainty with respect to the distribution of generated explanations for an answer.
  arXiv  Detail & Related papers  (2024-06-05T16:35:30Z)
• Model-Based Epistemic Variance of Values for Risk-Aware Policy Optimization [59.758009422067]
  We consider the problem of quantifying uncertainty over expected cumulative rewards in model-based reinforcement learning. We propose a new uncertainty Bellman equation (UBE) whose solution converges to the true posterior variance over values. We introduce a general-purpose policy optimization algorithm, Q-Uncertainty Soft Actor-Critic (QU-SAC) that can be applied for either risk-seeking or risk-averse policy optimization.
  arXiv  Detail & Related papers  (2023-12-07T15:55:58Z)
• 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)
• Integrating Uncertainty into Neural Network-based Speech Enhancement [27.868722093985006]
  Supervised masking approaches in the time-frequency domain aim to employ deep neural networks to estimate a multiplicative mask to extract clean speech. This leads to a single estimate for each input without any guarantees or measures of reliability. We study the benefits of modeling uncertainty in clean speech estimation.
  arXiv  Detail & Related papers  (2023-05-15T15:55:12Z)
• Optimizing the Performative Risk under Weak Convexity Assumptions [0.0]
  In performative prediction, a predictive model impacts the distribution that generates future data. Prior work has identified a pair of general conditions on the loss and the mapping from model parameters to distributions that implies convexity the performative risk. In this paper, we relax these assumptions, without sacrificing the amenability of the performative minimization risk problem for iterative optimization methods.
  arXiv  Detail & Related papers  (2022-09-02T01:07:09Z)
• A Geometric Method for Improved Uncertainty Estimation in Real-time [13.588210692213568]
  Post-hoc model calibrations can improve models' uncertainty estimations without the need for retraining. Our work puts forward a geometric-based approach for uncertainty estimation. We show that our method yields better uncertainty estimations than recently proposed approaches.
  arXiv  Detail & Related papers  (2022-06-23T09:18:05Z)
• Dense Uncertainty Estimation via an Ensemble-based Conditional Latent Variable Model [68.34559610536614]
  We argue that the aleatoric uncertainty is an inherent attribute of the data and can only be correctly estimated with an unbiased oracle model. We propose a new sampling and selection strategy at train time to approximate the oracle model for aleatoric uncertainty estimation. Our results show that our solution achieves both accurate deterministic results and reliable uncertainty estimation.
  arXiv  Detail & Related papers  (2021-11-22T08:54:10Z)
• Thompson Sampling Algorithms for Mean-Variance Bandits [97.43678751629189]
  We develop Thompson Sampling-style algorithms for mean-variance MAB. We also provide comprehensive regret analyses for Gaussian and Bernoulli bandits. Our algorithms significantly outperform existing LCB-based algorithms for all risk tolerances.
  arXiv  Detail & Related papers  (2020-02-01T15:33:50Z)

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.