Can News Predict the Market? Limits of Zero-Shot Financial NLP and the Role of Explainable AI

2026-06-10

AI Summary

This study investigates whether financial news can reliably predict short-term stock price movements without domain-specific training. To this end, the authors propose the first multi-level interpretable AI framework tailored for zero-shot financial NLP, integrating zero-shot natural language inference with time-weighted information fusion to explicitly model the recency and persistence of news events. The framework generates natural language rationales at three levels—token, article, and aggregate evidence—to support its predictions. Experimental results show that, overall, zero-shot approaches struggle to outperform simple baselines, particularly in forecasting price declines. Nevertheless, the interpretability signals produced by the model effectively identify prediction reliability, offering practical utility even under limited accuracy.

Abstract

Can financial news reliably predict short-term stock movements? Despite advances in large language models, this question remains unresolved. We revisit this problem using a zero-shot natural language processing framework, investigating whether models can extract actionable signals from financial news without domain-specific training. We design a structured pipeline that combines zero-shot natural language inference with temporal aggregation, explicitly modelling recency and event-dependent impact horizons when integrating information across articles. To address the need for transparency in high-stakes settings, we introduce a multi-layered explainability framework that links predictions to token-level, article-level, and aggregate evidence, and produces grounded natural language rationales. Across multiple models and prediction horizons, we find that zero-shot approaches consistently fail to outperform simple baselines, with particularly weak performance on negative movements, suggesting deeper structural limitations in mapping news sentiment to short-term price dynamics. However, explainability signals reliably distinguish between trustworthy and unreliable predictions, offering practical value even when accuracy is limited. These findings highlight the limits of zero-shot financial NLP and motivate a shift toward decision-support systems that prioritise transparency and uncertainty awareness. Code: https://github.com/alimert05/zero-shot-stock-xai

Problem

• financial news
• stock prediction
• zero-shot NLP
• explainable AI
• short-term price dynamics

Innovation

• zero-shot NLP
• explainable AI
• financial news prediction
• temporal aggregation
• decision-support systems