Detail publikačního výsledku

Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation

ČEGIŇ, J.; PECHER, B.; ŠIMKO, J.; SRBA, I.; BIELIKOVÁ, M.; BRUSILOVSKY, P.

Originální název

Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation

Anglický název

Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation

Druh

Stať ve sborníku mimo WoS a Scopus

Originální abstrakt

The generative large language models (LLMs) are increasingly used for data augmentation tasks, where text samples are paraphrased (or generated anew) and then used for classifier fine-tuning. Existing works on augmentation leverage the few-shot scenarios, where samples are given to LLMs as part of prompts, leading to better augmentations. Yet, the samples are mostly selected randomly and a comprehensive overview of the effects of other (more 'informed') sample selection strategies is lacking. In this work, we compare sample selection strategies existing in few-shot learning literature and investigate their effects in LLM-based textual augmentation. We evaluate this on in-distribution and out-of-distribution classifier performance. Results indicate, that while some 'informed' selection strategies increase the performance of models, especially for out-of-distribution data, it happens only seldom and with marginal performance increases. Unless further advances are made, a default of random sample selection remains a good option for augmentation practitioners.

Anglický abstrakt

The generative large language models (LLMs) are increasingly used for data augmentation tasks, where text samples are paraphrased (or generated anew) and then used for classifier fine-tuning. Existing works on augmentation leverage the few-shot scenarios, where samples are given to LLMs as part of prompts, leading to better augmentations. Yet, the samples are mostly selected randomly and a comprehensive overview of the effects of other (more 'informed') sample selection strategies is lacking. In this work, we compare sample selection strategies existing in few-shot learning literature and investigate their effects in LLM-based textual augmentation. We evaluate this on in-distribution and out-of-distribution classifier performance. Results indicate, that while some 'informed' selection strategies increase the performance of models, especially for out-of-distribution data, it happens only seldom and with marginal performance increases. Unless further advances are made, a default of random sample selection remains a good option for augmentation practitioners.

Klíčová slova

data augmentation, analysis

Klíčová slova v angličtině

data augmentation, analysis

Autoři

ČEGIŇ, J.; PECHER, B.; ŠIMKO, J.; SRBA, I.; BIELIKOVÁ, M.; BRUSILOVSKY, P.

Rok RIV

2026

Vydáno

04.11.2025

Nakladatel

Association for Computational Linguistics

Místo

Suzhou, China

ISBN

979-8-89176-335-7

Strany od

5533

Strany do

5550

Strany počet

18

URL

https://aclanthology.org/2025.findings-emnlp.296/

BibTex

@inproceedings{BUT193746,
  author="Ján {Čegiň} and Branislav {Pecher} and Jakub {Šimko} and  {} and Mária {Bieliková} and  {}",
  title="Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation",
  year="2025",
  pages="5533--5550",
  publisher="Association for Computational Linguistics",
  address="Suzhou, China",
  doi="10.18653/v1/2025.findings-emnlp.296",
  isbn="979-8-89176-335-7",
  url="https://aclanthology.org/2025.findings-emnlp.296/"
}