{"ID":2822731,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2601.06147","arxiv_id":"2601.06147","title":"LLM Flow Processes for Text-Conditioned Regression","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 \u003c ~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.","short_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 \u003c ~100 point...","url_abs":"https://arxiv.org/abs/2601.06147","url_pdf":"https://arxiv.org/pdf/2601.06147v2","authors":"[\"Felix Biggs\",\"Samuel Willis\"]","published":"2026-01-05T21:20:38Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.CL\",\"stat.ML\"]","methods":"[\"Diffusion Model\",\"Large Language Model\"]","has_code":false}