Related papers: Conformal Language Modeling via Posterior Sampling

Conformal Language Modeling via Posterior Sampling

URL: http://arxiv.org/abs/2606.03731v1
Date: Tue, 02 Jun 2026 14:49:15 GMT
Title: Conformal Language Modeling via Posterior Sampling
Abstract summary: Large Language Models remain plagued by hallucinations.<n>Recent work has sought to tame their prevalence using statistical techniques based on conformal prediction.<n>We propose to instead sample from approximations to an LLM posterior, where the conditioning event corresponds to a calibrated, high-scoring region.
Score: 12.200201135259817
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Large Language Models remain plagued by hallucinations. Recent work has sought to tame their prevalence using statistical techniques based on conformal prediction, with both theoretical and empirical success. However, these methods operate in a post-hoc fashion, treating the sampling procedure itself as atomic and then surgically altering samples to remove hallucinated claims. This disconnect between filtering and generation can result in samples that are incoherent, inconsistent, or simply unlikely under the model itself. Moreover, post-hoc surgery is unable to shift probability mass towards more useful and helpful responses. To address these issues, we propose to instead sample from approximations to an LLM posterior, where the conditioning event corresponds to a calibrated, high-scoring region. We develop a calibration procedure tailored to the setting of conditional sequential generation that effectively identifies this region and achieves target risk control. Empirically, we apply our method to case studies focused on open-ended biography generation and mathematical problem solving; compared to prior work, we obtain the same statistical guarantees, with higher downstream utility.

Related papers

Chance-constrained Flow Matching for High-Fidelity Constraint-aware Generation [46.932479632530764]
Chance-constrained Flow Matching integrates optimization into the sampling process, enabling effective enforcement of hard constraints.<n>Experiments show that CCFM outperforms current state-of-the-art constrained generative models in modeling complex physical systems.
arXiv Detail & Related papers (2025-09-29T17:56:52Z)
Counterfactual Generative Models for Time-Varying Treatments [15.208067770012283]
Estimating the counterfactual outcome of treatment is essential for decision-making in public health and clinical science. We propose a novel conditional generative framework capable of producing counterfactual samples under time-varying treatment. We present a thorough evaluation of our method using both synthetic and real-world data.
arXiv Detail & Related papers (2023-05-25T05:45:53Z)
Self-Normalized Importance Sampling for Neural Language Modeling [97.96857871187052]
In this work, we propose self-normalized importance sampling. Compared to our previous work, the criteria considered in this work are self-normalized and there is no need to further conduct a correction step. We show that our proposed self-normalized importance sampling is competitive in both research-oriented and production-oriented automatic speech recognition tasks.
arXiv Detail & Related papers (2021-11-11T16:57:53Z)
Mitigating Catastrophic Forgetting in Scheduled Sampling with Elastic Weight Consolidation in Neural Machine Translation [15.581515781839656]
Autoregressive models trained with maximum likelihood estimation suffer from exposure bias. We propose using Elastic Weight Consolidation as trade-off between mitigating exposure bias and retaining output quality. Experiments on two IWSLT'14 translation tasks demonstrate that our approach alleviates catastrophic forgetting and significantly improves BLEU.
arXiv Detail & Related papers (2021-09-13T20:37:58Z)
Scalable Marginal Likelihood Estimation for Model Selection in Deep Learning [78.83598532168256]
Marginal-likelihood based model-selection is rarely used in deep learning due to estimation difficulties. Our work shows that marginal likelihoods can improve generalization and be useful when validation data is unavailable.
arXiv Detail & Related papers (2021-04-11T09:50:24Z)
Bayesian prognostic covariate adjustment [59.75318183140857]
Historical data about disease outcomes can be integrated into the analysis of clinical trials in many ways. We build on existing literature that uses prognostic scores from a predictive model to increase the efficiency of treatment effect estimates.
arXiv Detail & Related papers (2020-12-24T05:19:03Z)
Increasing the efficiency of randomized trial estimates via linear adjustment for a prognostic score [59.75318183140857]
Estimating causal effects from randomized experiments is central to clinical research. Most methods for historical borrowing achieve reductions in variance by sacrificing strict type-I error rate control.
arXiv Detail & Related papers (2020-12-17T21:10:10Z)
Tracking disease outbreaks from sparse data with Bayesian inference [55.82986443159948]
The COVID-19 pandemic provides new motivation for estimating the empirical rate of transmission during an outbreak. Standard methods struggle to accommodate the partial observability and sparse data common at finer scales. We propose a Bayesian framework which accommodates partial observability in a principled manner.
arXiv Detail & Related papers (2020-09-12T20:37:33Z)
Improving Maximum Likelihood Training for Text Generation with Density Ratio Estimation [51.091890311312085]
We propose a new training scheme for auto-regressive sequence generative models, which is effective and stable when operating at large sample space encountered in text generation. Our method stably outperforms Maximum Likelihood Estimation and other state-of-the-art sequence generative models in terms of both quality and diversity.
arXiv Detail & Related papers (2020-07-12T15:31:24Z)

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