Related Experiment Video
Updated: Aug 6, 2026
08:19
Subnanometer-Resolution Structural Determination of Hemagglutinin from Cryo-Electron Tomography of Influenza Viruses
Published on: November 7, 2025
Fine-tuned large language models enhance influenza forecasting
Chenxiang Li1, Wenjing Gao1, Qiqiao Zhang1
1Department of Biostatistics and Systems Biology, School of Public Health (Shenzhen), Sun Yat-sen University, Shenzhen, Guangdong 510275, China.
Cell Reports Methods
|July 23, 2026
Summary
Fine-tuned large language models (LLMs) excel at influenza forecasting, even with limited data. These models offer accurate and stable predictions for public health surveillance, outperforming traditional methods.
Area of Science:
- Computational epidemiology
- Machine learning in public health
- Time-series forecasting
Background:
- Influenza-like illness (ILI) surveillance is crucial for public health.
- Accurate forecasting is needed for timely interventions.
- Data-limited settings pose challenges for traditional forecasting models.
Purpose of the Study:
- To benchmark fine-tuned large language models (LLMs) for influenza surveillance forecasting.
- To evaluate LLMs in data-limited, time-series settings.
- To compare LLM performance against established forecasting methods.
Main Methods:
- Developed a lightweight fine-tuning framework for pre-trained LLMs (Llama2, GPT2).
- Adapted LLMs with compact embedding and prediction layers.
- Evaluated models on seven weekly aggregated real-world surveillance datasets.
Main Results:
- Fine-tuned LLMs consistently outperformed SARIMA, LSTM, PatchTST, CoVTransformer, FEDformer, Time-LLM, and GPT4TS in accuracy and stability.
- LLMs showed superior performance in long-term forecasts across diverse regions.
- Pre-trained LLMs demonstrated competitive performance in zero-shot settings, capturing epidemic trends.
Conclusions:
- Fine-tuned LLMs are efficient and robust forecasting tools for public health surveillance.
- LLMs are suitable for privacy-sensitive and data-scarce applications.
- This approach enhances the capability for accurate influenza outbreak prediction.
Related Concept Videos
01:27Influenza
Influenza is an acute, highly communicable viral disease that affects the respiratory tract and is responsible for seasonal epidemics worldwide. Influenza A is the most prevalent type associated with widespread outbreaks and is subtyped based on two surface glycoproteins: hemagglutinin (H) and neuraminidase (N), as in H1N1. These glycoproteins are essential for viral infectivity, transmission, and immune recognition. Transmission occurs primarily through respiratory droplets and contaminated...
01:18Steps in Outbreak Investigation
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis: