uTeBC-NLP at SemEval-2024 Task 9: Can LLMs be Lateral Thinkers?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sadeghi, Pouya, Abaskohi, Amirhossein, Yaghoobzadeh, Yadollah
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929301968912384
author Sadeghi, Pouya
Abaskohi, Amirhossein
Yaghoobzadeh, Yadollah
author_facet Sadeghi, Pouya
Abaskohi, Amirhossein
Yaghoobzadeh, Yadollah
contents Inspired by human cognition, Jiang et al.(2023c) create a benchmark for assessing LLMs' lateral thinking-thinking outside the box. Building upon this benchmark, we investigate how different prompting methods enhance LLMs' performance on this task to reveal their inherent power for outside-the-box thinking ability. Through participating in SemEval-2024, task 9, Sentence Puzzle sub-task, we explore prompt engineering methods: chain of thoughts (CoT) and direct prompting, enhancing with informative descriptions, and employing contextualizing prompts using a retrieval augmented generation (RAG) pipeline. Our experiments involve three LLMs including GPT-3.5, GPT-4, and Zephyr-7B-beta. We generate a dataset of thinking paths between riddles and options using GPT-4, validated by humans for quality. Findings indicate that compressed informative prompts enhance performance. Dynamic in-context learning enhances model performance significantly. Furthermore, fine-tuning Zephyr on our dataset enhances performance across other commonsense datasets, underscoring the value of innovative thinking.
format Preprint
id arxiv_https___arxiv_org_abs_2404_02474
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle uTeBC-NLP at SemEval-2024 Task 9: Can LLMs be Lateral Thinkers?
Sadeghi, Pouya
Abaskohi, Amirhossein
Yaghoobzadeh, Yadollah
Computation and Language
Artificial Intelligence
Information Retrieval
Machine Learning
Inspired by human cognition, Jiang et al.(2023c) create a benchmark for assessing LLMs' lateral thinking-thinking outside the box. Building upon this benchmark, we investigate how different prompting methods enhance LLMs' performance on this task to reveal their inherent power for outside-the-box thinking ability. Through participating in SemEval-2024, task 9, Sentence Puzzle sub-task, we explore prompt engineering methods: chain of thoughts (CoT) and direct prompting, enhancing with informative descriptions, and employing contextualizing prompts using a retrieval augmented generation (RAG) pipeline. Our experiments involve three LLMs including GPT-3.5, GPT-4, and Zephyr-7B-beta. We generate a dataset of thinking paths between riddles and options using GPT-4, validated by humans for quality. Findings indicate that compressed informative prompts enhance performance. Dynamic in-context learning enhances model performance significantly. Furthermore, fine-tuning Zephyr on our dataset enhances performance across other commonsense datasets, underscoring the value of innovative thinking.
title uTeBC-NLP at SemEval-2024 Task 9: Can LLMs be Lateral Thinkers?
topic Computation and Language
Artificial Intelligence
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2404.02474