Argument Mining in BioMedicine: Zero-Shot, In-Context Learning and Fine-tuning with LLMs

Autoři: Jérémie Cabessa
Rok vydání: 2024
Typ publikace: Konferenční příspěvek (mezinárodní konf.)
Periodikum: Proceedings of the Tenth Italian Conference on Computational Linguistics (CLiC-it 2024)
Projekty: AppNeCo: Aproximativní neurovýpočty
Bibliografická citace: CABESSA, Jérémie, HERNAULT, H., MUSHTAQ, U. Argument Mining in BioMedicine: Zero-Shot, In-Context Learning and Fine-tuning with LLMs. In: DELL'ORLETTA, F., LENCI, A., MONTEMAGNI, S., SPRUGNOLI, R., eds. Proceedings of the Tenth Italian Conference on Computational Linguistics (CLiC-it 2024). Aachen: Technical University & CreateSpace Independent Publishing, 2024, č. článku 15. CEUR Workshop Proceedings, 3878. ISSN 1613-0073. Dostupné z: https://ceur-ws.org/Vol-3878/15_main_long.pdf
ID: 602747
Online záznam: Zobrazit online záznam

Abstrakt

Argument Mining (AM) aims to extract the complex argumentative structure of a text and Argument Type Classification (ATC) is an essential sub-task of AM. Large Language Models (LLMs) have shown impressive capabilities in most NLP tasks and beyond. However, fine-tuning LLMs can be challenging. In-Context Learning (ICL) has been suggested as a bridging paradigm between training-free and fine-tuning settings for LLMs. In ICL, an LLM is conditioned to solve tasks using a few solved demonstration examples included in its prompt. We focuse on AM in the biomedical AbstRCT dataset. We address ATC using quantized and unquantized LLaMA-3 models through zero-shot learning, in-context learning, and fine-tuning approaches. We introduce a novel ICL strategy that combines kNN-based example selection with majority vote ensembling, along with a well-designed fine-tuning strategy for ATC. In zero-shot setting, we show that LLaMA-3 fails to achieve acceptable classification results, suggesting the need for additional training modalities. However, in our ICL training-free setting, LLaMA-3 can leverage relevant information from only a few demonstration examples to achieve very competitive results. Finally, in our fine-tuning setting, LLaMA-3 achieves state-of-the-art performance on ATC task in AbstRCT dataset.