PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 1, 20260 citationsOpen Access

TMAK at NTCIR-18 FinArg-2 Task

TKTakahiro KawamotoXKXin Kang

Key Points

  • The aim is to classify text into premises and claims and assess their temporal relationships.
  • Participated in the FinArg-2 task
  • Classified text into 'Premise' and 'Claim' types
  • Trained a classification model on labeled data
  • Compared traditional machine learning approaches and large-scale language models
  • DeBERTa and Llama models achieved the highest classification accuracy
  • Large-scale language models outperformed traditional methods in classifying temporal relations

Abstract

This paper presents our participation in FinArg-2, which succceeds the FinArg-1 task. While FinArg-1 focused on sentiment analysis and argument classification, FinArg-2 extends this to temporal. We experiment with a method of classifying text into two types: "Premise" and "Claim." Based on these premises and claim, we have developed a method suitable for accurately classifying the temporal relationships between sentences. In order to classify sentences, we trained a classification model on labeled data, and compared traditional machine learning approaches with models that use large scale language models. Among the models tested, DeBERTa and Llama achieved the highest classification accuracy, demonstrating the model that used a large-scale language model showed auperior results.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Kawamoto et al. (2025) studied this question.

synapsesocial.com/papers/69cd7ad45652765b073a847ahttps://doi.org/10.20736/0002002040
Ask AI
Helpful
Bookmark
Share
View Full Paper