What question did this study set out to answer?

The research aims to improve the comprehension of financial arguments from various text sources to aid investors.

synapse

⌘+K

synapse

⌘+K

April 1, 2026Open Access

LIPI at the NTCIR-17 FinArg-1 Task: Using Pre-trained Language Models for Comprehending Financial Arguments

Key Points

The research aims to improve the comprehension of financial arguments from various text sources to aid investors.
Utilized pre-trained language models like BERT-SEC and FinBERT.
Employed cross-encoder architecture for processing financial texts.
Addressed three problems: Argument Unit Classification, Argument Relation Detection, and Identity of Argument Relations.
Achieved best performance in the NTCIR-17 FinArg tasks.
Successfully classified financial argument units.
Effectively detected and classified argument relations between financial texts.

Abstract

Comprehending arguments from financial texts helps investors in making data driven decisions. The FinArg tasks of NTCIR-17 deal with mining arguments related to finance from Research Reports, Earnings Conference Calls, and Social Media. In this paper, we describe our team's approach to solve the three such problems - Argument Unit Classification, Argument Relation Detection & Classification, and Identifying Attack and Support Argumentative Relations. We obtained best performance using pre-trained language models (like BERT-SEC and FinBERT) and cross-encoder architecture.

Mark Helpful

Bookmark

Relay

View Full Paper