Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jeong, Yujin, Jung, Noelle, Leung, Brian Y. C.
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.30652
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913172010565632
author Jeong, Yujin
Jung, Noelle
Leung, Brian Y. C.
author_facet Jeong, Yujin
Jung, Noelle
Leung, Brian Y. C.
contents Traditional multi-modal financial forecasting often relies on scalar sentiment scores, which fail to capture the nuances of financial news. To address this information loss, this paper explores high-dimensional representation learning by replacing discrete polarity ratings with dense FinBERT embeddings within a Transformer-based forecasting architecture. We benchmarked various embedding strategies on the FNSPID dataset, including raw embeddings, attention-weighted aggregation, and a custom Siamese network. While the attention-based mechanism struggled with the low signal-to-noise ratio typical of financial data, the integration of Siamese-optimized embeddings outperformed both the scalar baseline and raw embedding approaches, demonstrating that preserving high-dimensional narrative context yields improved predictive accuracy for short-term stock price movements.
format Preprint
id arxiv_https___arxiv_org_abs_2605_30652
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bridging the Gap Between Natural Language and Market Dynamics via High-Dimensional Representation Learning
Jeong, Yujin
Jung, Noelle
Leung, Brian Y. C.
Machine Learning
Traditional multi-modal financial forecasting often relies on scalar sentiment scores, which fail to capture the nuances of financial news. To address this information loss, this paper explores high-dimensional representation learning by replacing discrete polarity ratings with dense FinBERT embeddings within a Transformer-based forecasting architecture. We benchmarked various embedding strategies on the FNSPID dataset, including raw embeddings, attention-weighted aggregation, and a custom Siamese network. While the attention-based mechanism struggled with the low signal-to-noise ratio typical of financial data, the integration of Siamese-optimized embeddings outperformed both the scalar baseline and raw embedding approaches, demonstrating that preserving high-dimensional narrative context yields improved predictive accuracy for short-term stock price movements.
title Bridging the Gap Between Natural Language and Market Dynamics via High-Dimensional Representation Learning
topic Machine Learning
url https://arxiv.org/abs/2605.30652