Fine-tuned FLAN-T5-Large achieved the top METEOR score for claim extraction, while prompting methods scored lower but produced claims that human reviewers found more useful in some cases.
URL: https://huggingface.co/deepseek-ai/ DeepSeek-R1-Distill-Llama-8B
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
UNH at CheckThat! 2025: Fine-tuning Vs Prompting in Claim Extraction
Fine-tuned FLAN-T5-Large achieved the top METEOR score for claim extraction, while prompting methods scored lower but produced claims that human reviewers found more useful in some cases.