Saved in:
Bibliographic Details
Main Authors: Biggs, Felix, Willis, Samuel
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.06147
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913121206009856
author Biggs, Felix
Willis, Samuel
author_facet Biggs, Felix
Willis, Samuel
contents Recent work has demonstrated surprisingly good performance of pre-trained LLMs on regression tasks (for example, time-series prediction), with the ability to incorporate expert prior knowledge and the information contained in textual metadata. However we observe major error cascades even in short sequences < ~100 points; these models are also computationally intensive and difficult to parallelise. Marginal LLM predictions do not suffer this issue and are trivially parallelised, but can predict over-broad densities. To address this, we propose combining these densities with a lightweight (diffusion-based) neural process. We show that this combination leads to better-calibrated predictions overall, outputs locally consistent trajectories, and leads to text-conditioned function space selection in the meta-learner. As part of this work we propose a gradient-free (and non-Monte Carlo) method for sampling from a product-of-experts of a score model and an 'expert' (here the LLM predictive densities). We believe this general method is of independent interest as it is applicable whenever an expert can be convolved with a Gaussian in closed form.
format Preprint
id arxiv_https___arxiv_org_abs_2601_06147
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle LLM Flow Processes for Text-Conditioned Regression
Biggs, Felix
Willis, Samuel
Machine Learning
Computation and Language
Recent work has demonstrated surprisingly good performance of pre-trained LLMs on regression tasks (for example, time-series prediction), with the ability to incorporate expert prior knowledge and the information contained in textual metadata. However we observe major error cascades even in short sequences < ~100 points; these models are also computationally intensive and difficult to parallelise. Marginal LLM predictions do not suffer this issue and are trivially parallelised, but can predict over-broad densities. To address this, we propose combining these densities with a lightweight (diffusion-based) neural process. We show that this combination leads to better-calibrated predictions overall, outputs locally consistent trajectories, and leads to text-conditioned function space selection in the meta-learner. As part of this work we propose a gradient-free (and non-Monte Carlo) method for sampling from a product-of-experts of a score model and an 'expert' (here the LLM predictive densities). We believe this general method is of independent interest as it is applicable whenever an expert can be convolved with a Gaussian in closed form.
title LLM Flow Processes for Text-Conditioned Regression
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2601.06147