Evaluating Semantic Insensitivity in Financial Sentiment Analysis: Implications for Sustainable Financial Decision-Making
Summary
这篇论文提出了由LLM多智能体驱动的语义反转攻击框架ASI,对FinBERT等金融情感模型进行对抗性测试,发现其在90.85%的有效案例中被成功欺骗,揭示了金融NLP模型过度依赖关键词而缺乏真正语义理解的脆弱性。
View Cached Full Text
Cached at: 11/19/26, 02:58 AM
# Evaluating Semantic Insensitivity in Financial Sentiment Analysis: Implications for Sustainable Financial Decision-Making Source: [https://arxiv.org/abs/2611.21653](https://arxiv.org/abs/2611.21653) [View PDF](https://arxiv.org/pdf/2611.21653) > Abstract:Financial sentiment analysis is critical for automated trading, portfolio management, and risk assessment, heavily relying on models like FinBERT\. Despite high benchmark accuracy, these models remain vulnerable to subtle adversarial text modifications, threatening the integrity of algorithmic decision\-making\. In this paper, we propose Agent\-driven Semantic Inversion \(ASI\), a novel adversarial attack framework targeting financial sentiment models through domain\-specific semantic manipulation\. Utilizing an LLM\-powered multi\-agent pipeline, our approach identifies eligible financial news sentences and applies targeted lexical substitutions\. We empirically evaluate ASI against FinBERT, one of the most widely deployed financial sentiment models\. Our results reveal a striking vulnerability: the attack successfully deceived FinBERT in 90\.85% of eligible cases, with the model frequently maintaining its original sentiment prediction despite the fundamental inversion of financial performance indicators\. Through qualitative analysis and token\-level investigation, we demonstrate that FinBERT critically over\-relies on contextual keywords and syntactic patterns rather than true logical comprehension of financial semantics\. Our findings highlight the urgent need for logic\-aware, robust financial NLP models to ensure the reliability of automated decision\-making systems\. ## Submission history From: Gabriela Cioboata \[[view email](https://arxiv.org/show-email/e1fa51f4/2611.21653)\] **\[v1\]**Sun, 18 Oct 2026 16:43:13 UTC \(427 KB\)
Similar Articles
Enhancing Financial Sentiment Analysis via Retrieval Augmented Large Language Models
This paper introduces a retrieval-augmented LLM framework for financial sentiment analysis, achieving 15-48% improvement in accuracy and F1 score over traditional models and LLMs like ChatGPT and LLaMA.
Did the LLM Answer the Question? A Semantic Reconstruction Game for Jailbreak Evaluation
该论文提出了一种名为"语义重构博弈"的免训练评估方法,用于更准确地评估 LLM 越狱攻击的成功率(ASR)。研究发现仅更换评判 LLM 就可导致 ASR 报告值相差高达 82 个百分点,而新方法通过将合规性判断分解为受限语义比较,显著降低了相对于人工标注的平均误差。
Bridging the Gap Between Natural Language and Market Dynamics via High-Dimensional Representation Learning
This paper explores replacing scalar sentiment scores with high-dimensional FinBERT embeddings in a Transformer-based architecture for short-term stock price prediction, showing improved accuracy with Siamese-optimized embeddings.
Recognition Is Not Grounding: Benchmarking Multilingual Financial Evidence Extraction
论文提出 MULTIFINGROUND 多语言金融基准,揭示语言模型虽能跨语言识别财务概念(准确率 97.3-99.4%),却在 26.6-41.9% 的情况下无法完整恢复对应的数值证据,凸显识别与证据落地之间的差距。
FinLLM-KG: Knowledge Graph-Enhanced LLMs for Financial Sentiment Analysis in Algorithmic Trading
FinLLM-KG 是一个模型无关的框架,将因果 LLM 与金融知识图谱结合,从 SEC 10-K 文件中注入公司级、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。