Tag
This paper introduces FinEvolveBench, a benchmark for financial sentiment prediction, and Tree-of-Experience (ToE), a structured experience-management method for LLM agents in low-repetition tasks with implicit rewards. Experiments show that ToE outperforms general-purpose experience mechanisms in such challenging settings.
This paper presents a framework for Arabic financial sentiment analysis using LLMs, tailored for the Saudi market, integrating news and social media data to capture investor sentiment.
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.