Janus-Q: End-to-End Event-Driven Trading via Hierarchical-Gated Reward Modeling

2026-02-23

AI Summary

This work proposes the first end-to-end event-driven trading framework to address the limitations of existing learning-based trading systems, which struggle to effectively leverage financial news due to the absence of large-scale, event-centric datasets and misalignment between language models and trading objectives. The authors construct a novel dataset comprising 62,400 annotated news articles that explicitly links events to market reactions. Building upon this resource, they integrate supervised fine-tuning with reinforcement learning and introduce a Hierarchical Gated Reward Model (HGRM) to explicitly balance multiple trading objectives. The proposed approach achieves over a 17.5% improvement in directional prediction accuracy and up to a 102.0% increase in Sharpe ratio, significantly outperforming both market benchmarks and large language model baselines, thereby enabling more consistent, interpretable, and profitable trading decisions.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Financial market movements are often driven by discrete financial events conveyed through news, whose impacts are heterogeneous, abrupt, and difficult to capture under purely numerical prediction objectives. These limitations have motivated growing interest in using textual information as the primary source of trading signals in learning-based systems. Two key challenges hinder existing approaches: (1) the absence of large-scale, event-centric datasets that jointly model news semantics and statistically grounded market reactions, and (2) the misalignment between language model reasoning and financially valid trading behavior under dynamic market conditions. To address these challenges, we propose Janus-Q, an end-to-end event-driven trading framework that elevates financial news events from auxiliary signals to primary decision units. Janus-Q unifies event-centric data construction and model optimization under a two-stage paradigm. Stage I focuses on event-centric data construction, building a large-scale financial news event dataset comprising 62,400 articles annotated with 10 fine-grained event types, associated stocks, sentiment labels, and event-driven cumulative abnormal return (CAR). Stage II performs decision-oriented fine-tuning, combining supervised learning with reinforcement learning guided by a Hierarchical Gated Reward Model (HGRM), which explicitly captures trade-offs among multiple trading objectives. Extensive experiments demonstrate that Janus-Q achieves more consistent, interpretable, and profitable trading decisions than market indices and LLM baselines, improving the Sharpe Ratio by up to 102.0% while increasing direction accuracy by over 17.5% compared to the strongest competing strategies.

Problem

• event-driven trading
• financial news
• market reaction
• language model alignment
• trading signals

Innovation

• event-driven trading
• hierarchical gated reward model
• financial news events
• cumulative abnormal return
• end-to-end reinforcement learning