FinAnchor: Aligned Multi-Model Representations for Financial Prediction

TL;DR

FinAnchor is a lightweight framework that aligns and aggregates embeddings from multiple LLMs without fine-tuning, improving financial prediction accuracy across various NLP tasks by addressing representation incompatibility.

Contribution

It introduces a novel anchoring approach that aligns heterogeneous model embeddings into a common space for enhanced financial prediction.

Findings

FinAnchor outperforms single-model baselines.

It surpasses standard ensemble methods.

The framework is effective across multiple NLP tasks.

Abstract

Financial prediction from long documents involves significant challenges, as actionable signals are often sparse and obscured by noise, and the optimal LLM for generating embeddings varies across tasks and time periods. In this paper, we propose FinAnchor(Financial Anchored Representations), a lightweight framework that integrates embeddings from multiple LLMs without fine-tuning the underlying models. FinAnchor addresses the incompatibility of feature spaces by selecting an anchor embedding space and learning linear mappings to align representations from other models into this anchor. These aligned features are then aggregated to form a unified representation for downstream prediction. Across multiple financial NLP tasks, FinAnchor consistently outperforms strong single-model baselines and standard ensemble methods, demonstrating the effectiveness of anchoring heterogeneous representations for robust financial prediction.