WebCryptoAgent: Agentic Crypto Trading with Web Informatics
🤖 AI Summary
This work addresses the challenge of effectively integrating multi-source heterogeneous information and market microstructure signals in cryptocurrency trading under extreme volatility, while simultaneously ensuring decision robustness and rapid response to abrupt price shocks. The authors propose a multi-agent trading framework in which modality-specific agents process unstructured web content, social sentiment, and structured market data separately, then fuse these inputs into a unified evidence document followed by confidence-calibrated reasoning. A decoupled control architecture is introduced to separate hourly-level strategic reasoning from second-level real-time risk management. The approach pioneers a modality decomposition and evidence fusion mechanism to suppress spurious correlations and achieves disentanglement between strategic planning and risk response. Empirical results demonstrate that the system significantly enhances trading stability, reduces ineffective trades, and improves resilience to tail risks in real-world cryptocurrency markets.
Application Category
📝 Abstract
Cryptocurrency trading increasingly depends on timely integration of heterogeneous web information and market microstructure signals to support short-horizon decision making under extreme volatility. However, existing trading systems struggle to jointly reason over noisy multi-source web evidence while maintaining robustness to rapid price shocks at sub-second timescales. The first challenge lies in synthesizing unstructured web content, social sentiment, and structured OHLCV signals into coherent and interpretable trading decisions without amplifying spurious correlations, while the second challenge concerns risk control, as slow deliberative reasoning pipelines are ill-suited for handling abrupt market shocks that require immediate defensive responses. To address these challenges, we propose WebCryptoAgent, an agentic trading framework that decomposes web-informed decision making into modality-specific agents and consolidates their outputs into a unified evidence document for confidence-calibrated reasoning. We further introduce a decoupled control architecture that separates strategic hourly reasoning from a real-time second-level risk model, enabling fast shock detection and protective intervention independent of the trading loop. Extensive experiments on real-world cryptocurrency markets demonstrate that WebCryptoAgent improves trading stability, reduces spurious activity, and enhances tail-risk handling compared to existing baselines. Code will be available at https://github.com/AIGeeksGroup/WebCryptoAgent.
Problem
Research questions and friction points this paper is trying to address.
cryptocurrency trading
web informatics
market microstructure
risk control
real-time decision making
Innovation
Methods, ideas, or system contributions that make the work stand out.
agentic trading
web informatics
decoupled control architecture
multi-modal reasoning
real-time risk modeling
🔎 Similar Papers
No similar papers found.
