FinLLM-KG：用于算法交易中金融情感分析的知识图谱增强型大语言模型

arXiv cs.CL 论文
knowledge-graph financial-sentiment llm algorithmic-trading explainability retrieval

摘要

FinLLM-KG 是一个模型无关的框架，将因果大语言模型与金融知识图谱相结合，从美国证券交易委员会（SEC）的 10-K 文件中注入公司层面、时间对齐的知识，以改进金融情绪分析与算法交易，同时提供可解释的预测依据。

arXiv:2610.36083v1 Announce Type: new Abstract: Recent advances in Generative AI have substantially improved financial sentiment analysis through post-trained financial large language models (LLMs). However, existing approaches remain confined to predicting the sentiment of each article in isolation and operate as black boxes, which limits contextual understanding and hinders deployment in high-stakes financial decision-making. To this end, we introduce FinLLM-KG, a model-agnostic framework that combines causal LLMs with a financial knowledge graph, in order to inject temporally grounded, company-level knowledge from SEC 10-K filings at inference time, without updating model weights. Within FinLLM-KG, context is supplied either as a natural-language company profile or as a small set of relevance-ranked factual triples. This enables sentiment predictions to account for broader firm-level information rather than the article alone, at almost zero computational cost and for any causal LLM. In addition, attributing each article-level sentiment score to individual sentences and to the retrieved knowledge-graph facts makes the evidence behind the prediction inspectable, thereby enhancing explainability and supporting human-in-the-loop review. Simulations demonstrate that FinLLM-KG consistently improves portfolio performance over the state-of-the-art baseline.
查看原文
查看缓存全文

缓存时间: 2026/10/28 07:41 

# FinLLM-KG: Knowledge Graph-Enhanced LLMs for Financial Sentiment Analysis in Algorithmic Trading
Source: [https://arxiv.org/abs/2610.36083](https://arxiv.org/abs/2610.36083)
[View PDF](https://arxiv.org/pdf/2610.36083)[HTML \(experimental\)](https://arxiv.org/html/2610.36083v1)

> Abstract:Recent advances in Generative AI have substantially improved financial sentiment analysis through post\-trained financial large language models \(LLMs\)\. However, existing approaches remain confined to predicting the sentiment of each article in isolation and operate as black boxes, which limits contextual understanding and hinders deployment in high\-stakes financial decision\-making\. To this end, we introduce FinLLM\-KG, a model\-agnostic framework that combines causal LLMs with a financial knowledge graph, in order to inject temporally grounded, company\-level knowledge from SEC 10\-K filings at inference time, without updating model weights\. Within FinLLM\-KG, context is supplied either as a natural\-language company profile or as a small set of relevance\-ranked factual triples\. This enables sentiment predictions to account for broader firm\-level information rather than the article alone, at almost zero computational cost and for any causal LLM\. In addition, attributing each article\-level sentiment score to individual sentences and to the retrieved knowledge\-graph facts makes the evidence behind the prediction inspectable, thereby enhancing explainability and supporting human\-in\-the\-loop review\. Simulations demonstrate that FinLLM\-KG consistently improves portfolio performance over the state\-of\-the\-art baseline\.

## Submission history

From: Giorgos Iacovides \[[view email](https://arxiv.org/show-email/d4456430/2610.36083)\] **\[v1\]**Mon, 26 Oct 2026 18:08:20 UTC \(851 KB\)

相似文章