An Adaptive Multi Agent Bitcoin Trading System
Aadi Singhi
TL;DR
The paper addresses the challenge of forecasting and trading Bitcoin in a highly volatile, sentiment-driven market where traditional static models underperform. It proposes an adaptive, modular multi-agent system where LLMs act as specialized experts for quantitative analysis, sentiment assessment, and decision making, all guided by a dual verbal feedback mechanism that injects natural language critiques into prompts rather than retraining. Key contributions include the Verbal Feedback framework (daily Reflect and weekly Long Term Reflect), a Bitcoin price driver framework that quantifies the impact of technical, sentiment, and on-chain signals, and a modular architecture that enables scalable, explainable AI trading without parameter updates. Back-testing on 2024-07 to 2025-04 shows consistent outperformance across regimes, with the quantitative agent excelling in bullish markets, the sentiment-driven agent gaining in sideways conditions, and weekly feedback substantially boosting performance and reducing drawdown, underscoring the practicality of verbal feedback as a low-cost optimization loop for financial goals.
Abstract
This paper presents a Multi Agent Bitcoin Trading system that utilizes Large Language Models (LLMs) for alpha generation and portfolio management in the cryptocurrencies market. Unlike equities, cryptocurrencies exhibit extreme volatility and are heavily influenced by rapidly shifting market sentiments and regulatory announcements, making them difficult to model using static regression models or neural networks trained solely on historical data. The proposed framework overcomes this by structuring LLMs into specialised agents for technical analysis, sentiment evaluation, decision-making, and performance reflection. The agents improve over time via a novel verbal feedback mechanism where a Reflect agent provides daily and weekly natural-language critiques of trading decisions. These textual evaluations are then injected into future prompts of the agents, allowing them to adjust allocation logic without weight updates or finetuning. Back-testing on Bitcoin price data from July 2024 to April 2025 shows consistent outperformance across market regimes: the Quantitative agent delivered over 30\% higher returns in bullish phases and 15\% overall gains versus buy-and-hold, while the sentiment-driven agent turned sideways markets from a small loss into a gain of over 100\%. Adding weekly feedback further improved total performance by 31\% and reduced bearish losses by 10\%. The results demonstrate that verbal feedback represents a new, scalable, and low-cost approach of tuning LLMs for financial goals.