Probabilistic Attribution For Large Language Models

• URL: http://arxiv.org/abs/2605.21726v1
• Date: Wed, 20 May 2026 20:36:26 GMT
• Title: Probabilistic Attribution For Large Language Models
• Abstract summary: The generative nature of Large Language Models (LLMs) is reflected in the conditional probabilities they compute to sample each response token given the previous tokens.<n>In this work, we use these probabilities to situate LLMs within the mathematical theory of processes.<n>We use this framework to design a model-agnostic probabilistic token attribution measure, using Bayes rule to invert the next-token log-probabilities.
• Score: 4.83127178873072
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The generative nature of Large Language Models (LLMs) is reflected in the conditional probabilities they compute to sample each response token given the previous tokens. These probabilities encode the distributional structure that the model learns in training and exploits in inference. In this work, we use these probabilities to situate LLMs within the mathematical theory of stochastic processes. We use this framework to design a model-agnostic probabilistic token attribution measure, using Bayes rule to invert the next-token log-probabilities so as to capture the models internal representation of the distribution over token sequences. The representation is independent of the models computational structure. This representation yields the conditional probability of the response given the prompt, and of the response given the prompt with a token marginalized away. Our attribution score is the log of the ratio of these probabilities. We further compute the entropies of a single prompts token distributions, conditioned on the remaining context. The interplay between entropy and attribution score sheds light on LLM behavior. We evaluate 8 models across 7 prompts and investigate anomalies, token sensitivity, response stability, model stability, and training convergence, thereby improving interpretability and guiding users to focus on uncertain or unstable parts of the generation.

Related papers

• Probabilistic classification from possibilistic data: computing Kullback-Leibler projection with a possibility distribution [12.607603625414571]
  We consider learning with possibilistic supervision for multi-class classification.<n>For each training instance, the supervision is a normalized possibility distribution.<n>We construct a non-empty closed convex set of admissible probability distributions.
  arXiv  Detail & Related papers  (2026-04-02T11:58:52Z)
• D-Models and E-Models: Diversity-Stability Trade-offs in the Sampling Behavior of Large Language Models [91.21455683212224]
  In large language models (LLMs), the probability of relevance for the next piece of information is linked to the probability of relevance for the next product.<n>But whether fine-grained sampling probabilities faithfully align with task requirements remains an open question.<n>We identify two model types: D-models, whose P_token exhibits large step-to-step variability and poor alignment with P_task; and E-models, whose P_token is more stable and better aligned with P_task.
  arXiv  Detail & Related papers  (2026-01-25T14:59:09Z)
• Certain but not Probable? Differentiating Certainty from Probability in LLM Token Outputs for Probabilistic Scenarios [1.1510009152620668]
  We investigate the relationship between token certainty and alignment with theoretical probability distributions in well-defined probabilistic scenarios.<n>We measure two dimensions: (1) response validity with respect to scenario constraints, and (2) alignment between token-level output probabilities and theoretical probabilities.<n>Our results indicate that, while both models achieve perfect in-domain response accuracy across all prompt scenarios, their token-level probability and entropy values consistently diverge from the corresponding theoretical distributions.
  arXiv  Detail & Related papers  (2025-11-01T16:51:11Z)
• 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)
• Estimating Causal Effects from Learned Causal Networks [56.14597641617531]
  We propose an alternative paradigm for answering causal-effect queries over discrete observable variables. We learn the causal Bayesian network and its confounding latent variables directly from the observational data. We show that this emphmodel completion learning approach can be more effective than estimand approaches.
  arXiv  Detail & Related papers  (2024-08-26T08:39:09Z)
• Do LLMs Play Dice? Exploring Probability Distribution Sampling in Large Language Models for Behavioral Simulation [73.58618024960968]
  An increasing number of studies are employing large language models (LLMs) as agents to emulate the sequential decision-making processes of humans.<n>This arouses curiosity regarding the capacity of LLM agents to comprehend probability distributions.<n>Our analysis indicates that LLM agents can understand probabilities, but they struggle with probability sampling.
  arXiv  Detail & Related papers  (2024-04-13T16:59:28Z)
• Logistic-beta processes for dependent random probabilities with beta marginals [58.91121576998588]
  We propose a novel process called the logistic-beta process, whose logistic transformation yields a process with common beta marginals.<n>We show how the proposed logistic-beta process can be used to design computationally tractable dependent nonparametric models.
  arXiv  Detail & Related papers  (2024-02-10T21:41:32Z)
• Self-Consistency of Large Language Models under Ambiguity [4.141513298907867]
  This work presents an evaluation benchmark for self-consistency in cases of under-specification. We conduct a series of behavioral experiments on the OpenAI model suite using an ambiguous integer sequence completion task. We find that average consistency ranges from 67% to 82%, far higher than would be predicted if a model's consistency was random.
  arXiv  Detail & Related papers  (2023-10-20T11:57:56Z)
• 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)
• Probabilistic Kolmogorov-Arnold Network [1.4732811715354455]
  The present paper proposes a method for estimating probability distributions of the outputs in the case of aleatoric uncertainty. The suggested approach covers input-dependent probability distributions of the outputs, as well as the variation of the distribution type with the inputs. Although the method is applicable to any regression model, the present paper combines it with KANs, since the specific structure of KANs leads to computationally-efficient models' construction.
  arXiv  Detail & Related papers  (2021-04-04T23:49:15Z)
• Tractable Inference in Credal Sentential Decision Diagrams [116.6516175350871]
  Probabilistic sentential decision diagrams are logic circuits where the inputs of disjunctive gates are annotated by probability values. We develop the credal sentential decision diagrams, a generalisation of their probabilistic counterpart that allows for replacing the local probabilities with credal sets of mass functions. For a first empirical validation, we consider a simple application based on noisy seven-segment display images.
  arXiv  Detail & Related papers  (2020-08-19T16:04:34Z)
• Structural Causal Models Are (Solvable by) Credal Networks [70.45873402967297]
  Causal inferences can be obtained by standard algorithms for the updating of credal nets. This contribution should be regarded as a systematic approach to represent structural causal models by credal networks. Experiments show that approximate algorithms for credal networks can immediately be used to do causal inference in real-size problems.
  arXiv  Detail & Related papers  (2020-08-02T11:19:36Z)

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.