Loading
Life sciences · Preprint
arXiv · August 12, 2026
Posted before peer review. The findings may change or fail to hold.
FM-LLM is a novel neural network architecture combining large language models with frequency-domain analysis for automated time-series forecasting. The method achieves reported improvements in standard numerical metrics (MSE and MAE) across 59 of 78 evaluation metrics on eleven public benchmarks, but the work is unpublished, lacks peer review, and does not address clinical or operational validation.
Preprint. Intervention: FM-LLM: a Fourier Analysis Network-based spectral token aligner with asymmetric Mixture-of-Experts decoder and time-frequency hybrid loss function for time-series forecasting using frozen large language models.. Compared with: Strongest autoregressive LLM-based baseline.
State-of-the-art performance on 59 out of 78 evaluation metrics across eleven public benchmarks Average improvement of 5.3% in MSE and 5.6% in MAE compared to strongest autoregressive LLM baseline Maximum gains reaching 8.0% for MSE and 8.4% for MAE
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
This is an unrefereed technical preprint describing a novel machine learning architecture for time-series forecasting; it reports computational results on benchmarks but has not undergone peer review.
As stated by the source record.
Quoted from the source exactly as published.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
Recent advances in Large Language Models (LLMs) have spurred cross-modal solutions for time-series forecasting. However, existing methods rely heavily on textual prompts for modality alignment-introducing nontrivial computational overhead and failing to leverage the rich spectral dynamics inherent in time-series data. To enable prompt-free, frequency-aware adaptation of frozen LLMs, we propose FM-LLM (Frequency-Enhanced Mixture-of-Experts for adapting LLMs to Time Series Forecasting), an autoregressive framework grounded in constrained asymmetric coupling. A Fourier Analysis Network (FAN)-based spectral token aligner injects structured harmonic representations directly into the frozen LLM with numerical compatibility. An asymmetric Mixture-of-Experts (MoE) decoder enforces role separation: shared experts with lightweight FAN layers reconstruct the global periodic backbone, while routed experts-restricted to standard FFNs-specialize in modeling non-periodic residual dynamics. A time-frequency hybrid loss function jointly optimizes temporal accuracy and spectral consistency, mitigating error accumulation during long-horizon autoregressive rollouts. Evaluated across eleven public benchmarks, FM-LLM achieves state-of-the-art performance on 59 out of 78 evaluation metrics. Compared to the strongest autoregressive LLM-based baseline, it delivers average improvements of 5.3% in MSE and 5.6% in MAE, with maximum gains reaching 8.0% for MSE and 8.4% for MAE. FM-LLM also demonstrates robust transferability, maintaining superior performance in 10% few-shot and zero-shot forecasting scenarios.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.