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

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F26%3A0193746" target="_blank" >RIV/00216305:26230/26:0193746 - isvavai.cz</a>

Výsledek na webu

<a href="https://aclanthology.org/2025.findings-emnlp.296/" target="_blank" >https://aclanthology.org/2025.findings-emnlp.296/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.18653/v1/2025.findings-emnlp.296" target="_blank" >10.18653/v1/2025.findings-emnlp.296</a>

Jazyk výsledku

angličtina

Název v původním jazyce

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

Popis výsledku v původním jazyce

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.

Název v anglickém jazyce

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

Popis výsledku anglicky

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.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů