Abstract: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.
| Subjects: | Machine Learning (cs.LG); Computation and Language (cs.CL); Machine Learning (stat.ML) |
| Cite as: | arXiv:2601.06147 [cs.LG] |
| (or arXiv:2601.06147v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2601.06147 arXiv-issued DOI via DataCite |
Submission history
From: Samuel Willis [view email]
[v1]
Mon, 5 Jan 2026 21:20:38 UTC (7,075 KB)
[v2]
Wed, 13 May 2026 11:02:21 UTC (755 KB)