English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Parameter efficient finetuning has emerged as a viable solution for improvingthe performance of Large Language Models without requiring massive resourcesand compute. Prior work on multilingual evaluation has shown that there is alarge gap between the performance of LLMs on English and other languages.Further, there is also a large gap between the performance of smalleropen-source models and larger LLMs. Finetuning can be an effective way tobridge this gap and make language models more equitable. In this work, wefinetune the LLaMA-7B and Mistral-7B models on synthetic multilingualinstruction tuning data to determine its effect on model performance on fivedownstream tasks covering twenty three languages in all. Additionally, weexperiment with various parameters, such as rank for low-rank adaptation andvalues of quantisation to determine their effects on downstream performance andfind that higher rank and higher quantisation values benefit low-resourcelanguages. We find that parameter efficient finetuning of smaller open sourcemodels sometimes bridges the gap between the performance of these models andthe larger ones, however, English performance can take a hit. We also find thatfinetuning sometimes improves performance on low-resource languages, whiledegrading performance on high-resource languages.

MAPLE: Multilingual Evaluation of Parameter Efficient Finetuning of Large Language Models