Beyond Sentiment: Structured Information Extraction from Financial News

Researchers from an unnamed institution proposed a structured information extraction framework using LLaMA-3.1-70B to extract six semantic dimensions from financial news, finding that combining these with FinBERT sentiment features improves stock movement prediction F1 to 0.600, significantly outperforming sentiment alone (p < 0.0001). The study, based on 41,618 news–stock pairs from the FNSPID dataset, shows that non-sentiment dimensions contribute an additional ΔF1 = +0.019 over FinBERT alone, indicating that sentiment-only analysis incurs substantial information loss.

arXiv:2607.28496v1 Announce Type: new Abstract: Financial sentiment analysis has become a standard component in news-driven stock prediction, yet it reduces rich, multi-dimensional news articles to a single polarity score. We hypothesize that financial news encodes multiple orthogonal information dimensions---event type, impact scope, temporal horizon, and semantic confidence---that sentiment alone cannot capture, and that these dimensions carry independent predictive value. To test this hypothesis, we propose a structured information extraction framework that leverages LLaMA-3.1-70B to extract six semantic dimensions from financial news. Through large-scale experiments on 41,618 news--stock pairs from the FNSPID dataset, we find that i FinBERT sentiment features exhibit strong predictive power under nonlinear models F1=0.576 but substantially weaker performance under linear models F1=0.230 , revealing a highly nonlinear sentiment--return relationship; ii LLM-extracted structured features, while individually weaker, capture information orthogonal to sentiment, as evidenced by a 53.5% systematic disagreement rate between the two approaches; and iii combining both signal sources yields F1=0.600, significantly outperforming either alone $p < 0.0001$ , with consistent improvements across all seven event types. Ablation experiments confirm that non-sentiment structural dimensions event type, impact subject, time horizon, confidence independently contribute $\Delta\text{F1} = +0.019$ beyond FinBERT alone. Feature importance analysis reveals balanced contributions from all six extracted dimensions 14--21% , demonstrating that compressing news into a single sentiment score incurs substantial information loss. Our results suggest that the sentiment--semantics decoupling in financial text is systematic and exploitable, opening a new direction for multi-dimensional financial NLP.