UNH at CheckThat! 2025: Fine-tuning Vs Prompting in Claim Extraction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wilder, Joe, Kadapala, Nikhil, Xu, Benji, Alsaadi, Mohammed, Parsons, Aiden, Rogers, Mitchell, Agarwal, Palash, Hassick, Adam, Dietz, Laura
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911142990839808
author Wilder, Joe
Kadapala, Nikhil
Xu, Benji
Alsaadi, Mohammed
Parsons, Aiden
Rogers, Mitchell
Agarwal, Palash
Hassick, Adam
Dietz, Laura
author_facet Wilder, Joe
Kadapala, Nikhil
Xu, Benji
Alsaadi, Mohammed
Parsons, Aiden
Rogers, Mitchell
Agarwal, Palash
Hassick, Adam
Dietz, Laura
contents We participate in CheckThat! Task 2 English and explore various methods of prompting and in-context learning, including few-shot prompting and fine-tuning with different LLM families, with the goal of extracting check-worthy claims from social media passages. Our best METEOR score is achieved by fine-tuning a FLAN-T5 model. However, we observe that higher-quality claims can sometimes be extracted using other methods, even when their METEOR scores are lower.
format Preprint
id arxiv_https___arxiv_org_abs_2509_06883
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UNH at CheckThat! 2025: Fine-tuning Vs Prompting in Claim Extraction
Wilder, Joe
Kadapala, Nikhil
Xu, Benji
Alsaadi, Mohammed
Parsons, Aiden
Rogers, Mitchell
Agarwal, Palash
Hassick, Adam
Dietz, Laura
Computation and Language
Artificial Intelligence
Information Retrieval
We participate in CheckThat! Task 2 English and explore various methods of prompting and in-context learning, including few-shot prompting and fine-tuning with different LLM families, with the goal of extracting check-worthy claims from social media passages. Our best METEOR score is achieved by fine-tuning a FLAN-T5 model. However, we observe that higher-quality claims can sometimes be extracted using other methods, even when their METEOR scores are lower.
title UNH at CheckThat! 2025: Fine-tuning Vs Prompting in Claim Extraction
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2509.06883