FinDialogLens: Event Extraction over Multi-Party Dialogue for Missed-Trade Identification in Financial Chatrooms

2026-10-01

AI Summary

This study addresses the challenge of manually detecting missed orders in multi-party financial chatrooms, where interleaved conversations complicate human oversight. We formulate missed order detection as an event extraction task over multi-party dialogues and propose a hybrid large language model (LLM) pipeline architecture. A novel difficulty-aware routing mechanism dynamically allocates tasks between rule-based engines and LLMs. Furthermore, RFQ-level modular segmentation combined with a Trade Engine enables precise extraction of inquiry triggers and execution outcomes. Experimental results demonstrate that our approach achieves 92.1% accuracy in price extraction and 94.3% in trade identification while reducing LLM invocations by 85%. This yields daily cost savings exceeding $300, effectively balancing high predictive accuracy with low inference overhead.

Abstract

Multi-party financial chatrooms are vital for sales-and-trading professionals, but their complexity makes manual recovery of missed trades infeasible: each Request for Quote (RFQ) is an event whose final price and trade outcome appear many messages after the RFQ-trigger message (the inquiry message), interleaved with concurrent RFQs from other participants. We cast this as event extraction (EE) over multi-party dialogue and present FinDialogLens, a hybrid LLM pipeline in which compact fine-tuned classifiers act as inference-time scaffolds: they detect RFQ-triggers and price/trade outcome metadata, an RFQ-Level Module segments per-event RFQ windows, and a Trade Engine fills argument roles. With GPT-4o, FinDialogLens reaches 92.1% and 94.3% accuracy on final price and trade outcome, respectively, outperforming full-chatroom CoT prompting methods; fine-tuned open-source LLMs with as few as 3B parameters achieve comparable performance with modest in-domain data. To make the LLM-based solution practical at scale, a difficulty-aware router balances cost and accuracy by allocating RFQs between a low-cost rule-based engine and the higher-performing LLM-powered Trade Engine, cutting LLM calls by 85% on final price while recovering half of the accuracy gap to FinDialogLens (GPT-4o), saving over $300/day at our 70,000-RFQ/day scale.

Problem

• Missed-Trade Identification
• Event Extraction
• Multi-Party Dialogue
• Financial Chatrooms
• Request for Quote

Innovation

• Event Extraction
• Multi-Party Dialogue
• Hybrid LLM Pipeline
• Difficulty-Aware Router
• Missed-Trade Identification