EF-LLM：基于持续学习的能源预测大语言模型框架，支持自动化、稀疏预测增强与幻觉检测
EF-LLM: A Continual Learning-Based Framework for AI-Assisted Automation, Enhanced Sparse Prediction, and Hallucination Detection
AI 导读
研究者提出面向能源预测的大语言模型 EF-LLM，通过持续学习框架、可更新的 LoRA 与多通道架构融合领域知识与时间序列数据，并用 F-PEFT（Fusion Parameter-Efficient Fine-Tuning）方法结合时序与文本数据，提升稀疏数据条件下的预测精度。
正文 · 原文
Abstract:Accurate prediction helps achieve supply-demand balance in energy systems,thereby supporting operational decision-making and this http URL models,which lack AI-assisted automation,often rely heavily on expert knowledge,incur high labor costs,and perform poorly under sparse-data this http URL address these challenges,this work proposes an Energy Forecasting Large Language Model (EF-LLM),which integrates domain knowledge and temporal data for time-series forecasting while supporting both pre-forecast operations and post-forecast this http URL-LLM's human-artificial intelligence (AI) interaction capabilities further lower the entry barrier for forecasting tasks and reduce the need for additional expert this http URL support these functions, a continual learning framework with updatable Low-Rank Adaptation (LoRA) and a multi-channel architecture is developed to align heterogeneous multimodal data, enabling EF-LLM to continuously acquire and adapt multimodal this http URL addition,EF-LLM improves prediction accuracy under sparse-data conditions by leveraging its ability to process and fuse multimodal information.A Fusion Parameter-Efficient Fine-Tuning (F-PEFT) method is further introduced to effectively exploit both time-series data and textual this http URL-LLM is also the first energy-specific LLM to detect hallucinations and quantify their occurrence rate, achieved via multi-task learning and semantic similarity this http URL work has achieved success in energy prediction scenarios for load, PV, and wind power forecast.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2411.00852 [cs.LG] |
| (or arXiv:2411.00852v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2411.00852 arXiv-issued DOI via DataCite |
Submission history
From: Zihang Qiu [view email]
[v1]
Wed, 30 Oct 2024 11:22:37 UTC (7,670 KB)
[v2]
Tue, 24 Dec 2024 03:24:55 UTC (38,545 KB)
[v3]
Mon, 26 Oct 2026 04:59:48 UTC (31,594 KB)
来源：arXiv cs.AI · arxiv.org