PumpSense: Real-Time Detection and Target Extraction of Crypto Pump-and-Dumps on Telegram

2026-05-10

AI Summary

This study addresses the critical threat posed by coordinated pump-and-dump manipulation of cryptocurrencies via Telegram, which undermines market integrity yet remains challenging to detect with both high accuracy and real-time responsiveness due to the scarcity of fine-grained annotated data. To bridge this gap, the authors construct the first large-scale Telegram dataset comprising 280,000 human-annotated messages and define two core tasks: real-time detection of pump-and-dump announcements and extraction of targeted cryptocurrencies and exchanges. They propose a novel detection framework integrating LightGBM with BGE-M3 embeddings, achieving an F1 score of 0.83 with a latency as low as 50 milliseconds. Furthermore, they introduce, for the first time, a large language model for entity extraction, attaining an accuracy of 0.91—significantly outperforming conventional rule-based approaches—and enabling microsecond-level early warnings without reliance on trading data.

Abstract

Cryptocurrency pump-and-dump schemes coordinated via Telegram threaten market integrity. However, existing research addressing this specific threat has not yet produced solutions that combine reliable results with fast response. This is in part due to the absence of publicly available, message-level labeled data, as well as design choices. In this paper, we address both issues. In particular, we introduce a corpus of over 280,000 Telegram posts from 39 pump-organizing groups, all manually reviewed to identify 2,246 pump announcements and their targeted cryptocurrency and exchange. Leveraging this dataset, we define two tasks: real-time pump-announcement detection and target cryptocurrency/exchange extraction. For detection, we compare two machine-learning models: a lightweight tree-based LightGBM classifier (F1=0.79, latency=9.4 s/sample) and a transformer-based BGE-M3 (F1=0.83, latency=50 ms/sample). With our proposed approach, we show that message analysis can achieve near-instant pump detection at the level of individual Telegram message windows. Unlike prior work that relies purely on market data and typically detects pumps tens of seconds after abnormal trading activity is observed, our method operates directly on the coordination messages themselves and can be evaluated in microseconds per window on commodity hardware. To our knowledge, we also establish the first benchmark for manipulated coin and exchange extraction. We demonstrate that traditional rule-based extraction methods, widely relied upon in prior literature, are ineffective due to ticker ambiguity. In contrast, LLMs achieve the highest accuracy with a score of 0.91.

Problem

• pump-and-dump
• cryptocurrency
• Telegram
• market manipulation
• real-time detection

Innovation

• pump-and-dump detection
• Telegram message analysis
• real-time cryptocurrency monitoring
• target extraction
• large language models