Salvato in:
Dettagli Bibliografici
Autore principale: Moell, Birger
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2405.14006
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911884806979584
author Moell, Birger
author_facet Moell, Birger
contents In the rapidly evolving field of artificial intelligence, large language models (LLMs) have demonstrated significant capabilities across numerous applications. However, the performance of these models in languages with fewer resources, such as Swedish, remains under-explored. This study introduces a comprehensive human benchmark to assess the efficacy of prominent LLMs in understanding and generating Swedish language texts using forced choice ranking. We employ a modified version of the ChatbotArena benchmark, incorporating human feedback to evaluate eleven different models, including GPT-4, GPT-3.5, various Claude and Llama models, and bespoke models like Dolphin-2.9-llama3b-8b-flashback and BeagleCatMunin. These models were chosen based on their performance on LMSYS chatbot arena and the Scandeval benchmarks. We release the chatbotarena.se benchmark as a tool to improve our understanding of language model performance in Swedish with the hopes that it will be widely used. We aim to create a leaderboard once sufficient data has been collected and analysed.
format Preprint
id arxiv_https___arxiv_org_abs_2405_14006
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Evaluating Large Language Models with Human Feedback: Establishing a Swedish Benchmark
Moell, Birger
Computation and Language
Artificial Intelligence
In the rapidly evolving field of artificial intelligence, large language models (LLMs) have demonstrated significant capabilities across numerous applications. However, the performance of these models in languages with fewer resources, such as Swedish, remains under-explored. This study introduces a comprehensive human benchmark to assess the efficacy of prominent LLMs in understanding and generating Swedish language texts using forced choice ranking. We employ a modified version of the ChatbotArena benchmark, incorporating human feedback to evaluate eleven different models, including GPT-4, GPT-3.5, various Claude and Llama models, and bespoke models like Dolphin-2.9-llama3b-8b-flashback and BeagleCatMunin. These models were chosen based on their performance on LMSYS chatbot arena and the Scandeval benchmarks. We release the chatbotarena.se benchmark as a tool to improve our understanding of language model performance in Swedish with the hopes that it will be widely used. We aim to create a leaderboard once sufficient data has been collected and analysed.
title Evaluating Large Language Models with Human Feedback: Establishing a Swedish Benchmark
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2405.14006