EXAONE Forecast for Finance
EXAONE Forecast for Finance
Abstract: This technical report presents EXAONE Forecast for Finance (EXAONE Finance), a financial time series (TS) foundation model (TSFM) tailored to financial forecasting. 摘要： 本技术报告介绍了 EXAONE Forecast for Finance (EXAONE Finance)，这是一款专为金融预测量身定制的金融时间序列 (TS) 基础模型 (TSFM)。
Recent TSFMs achieve strong zero-shot performance through large-scale pretraining. However, they are primarily developed for general-domain TS and largely rely on self-attention backbones whose computational cost grows quadratically with sequence length and variate count. 近期的 TSFM 通过大规模预训练实现了强大的零样本性能。然而，它们主要针对通用领域的时间序列开发，且很大程度上依赖于自注意力主干网络，其计算成本会随着序列长度和变量数量的增加呈二次方增长。
Moreover, they assume fully observed inputs and are pretrained on corpora that fail to capture the unique dynamics of financial markets. These limitations hinder their applicability to finance, where long, many-channel, intermittently observed panels are common. 此外，这些模型通常假设输入是完全观测到的，且其预训练语料库未能捕捉金融市场的独特动态。这些局限性阻碍了它们在金融领域的应用，因为金融数据往往具有长序列、多通道且观测间歇性的特点。
To address these challenges, EXAONE Finance adopts an attention-free architecture, replacing self-attention with two simple yet effective linear-time operators: 1) a causal 1D convolution for temporal mixing and 2) a group-aware pooling multi-layer perceptron (MLP) for variate mixing. 为了应对这些挑战，EXAONE Finance 采用了无注意力架构，用两个简单且有效的线性时间算子取代了自注意力机制：1) 用于时间混合的因果一维卷积；2) 用于变量混合的组感知池化多层感知机 (MLP)。
Furthermore, a masked context augmentation exposes the model to contiguous missing spans during training, improving robustness to the missingness pervasive in financial markets. 此外，掩码上下文增强技术在训练过程中让模型接触到连续的缺失片段，从而提高了模型对金融市场中普遍存在的缺失数据的鲁棒性。
EXAONE Finance is pretrained on a large-scale financial corpus covering not only equities but also foreign exchange, commodities, crypto-assets, fixed income, and macroeconomic indicators. EXAONE Finance 在大规模金融语料库上进行了预训练，涵盖了股票、外汇、大宗商品、加密资产、固定收益以及宏观经济指标等多个领域。
On FinVerse, a financial forecasting benchmark covering diverse asset classes, EXAONE Finance attains state-of-the-art performance, ranking first across all three evaluation tiers---point-forecast accuracy, cross-sectional asset ranking, and portfolio profitability. 在涵盖多种资产类别的金融预测基准 FinVerse 上，EXAONE Finance 达到了业界领先水平，在点预测准确性、横截面资产排名和投资组合盈利能力这三个评估维度上均排名第一。