Assessing and Enhancing the Robustness of Large Language Models with Task Structure Variations for Logical Reasoning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bao, Qiming, Gendron, Gael, Peng, Alex Yuxuan, Zhong, Wanjun, Tan, Neset, Chen, Yang, Witbrock, Michael, Liu, Jiamou
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916569082232832
author Bao, Qiming
Gendron, Gael
Peng, Alex Yuxuan
Zhong, Wanjun
Tan, Neset
Chen, Yang
Witbrock, Michael
Liu, Jiamou
author_facet Bao, Qiming
Gendron, Gael
Peng, Alex Yuxuan
Zhong, Wanjun
Tan, Neset
Chen, Yang
Witbrock, Michael
Liu, Jiamou
contents Large language models (LLMs), such as LLaMA, Alpaca, Vicuna, GPT-3.5 and GPT-4, have advanced the performance of AI systems on various natural language processing tasks to human-like levels. However, their generalisation and robustness when performing logical reasoning has not been sufficiently assessed. To comprehensively evaluate this ability, we develop three new logical reasoning datasets named "ReClor-plus", "LogiQA-plus" and "LogiQAv2-plus" that extend standard logical reasoning datasets to evaluate the robustness of the LLM's reasoning. For each, we create three subsets: the first with randomly shuffled options, the second with the correct choices replaced by "none of the other options is correct", and the third with a combination of shuffling and substitution. Experiments on these datasets show that these simple augmentations greatly hinder the models' performance. Despite their high performance on the original publicly available datasets, we find that all models perform poorly on these newly constructed datasets. We also demonstrate that introducing task variations into the training set can markedly improve the model's performance on both the original and our developed datasets. Finally, we show that applying logic-driven data augmentation for fine-tuning and prompting can enhance generalisation in both discriminative and generative models, offering a path to improving their robustness for tasks involving logical reasoning. Source code and data are made publicly available at https://github.com/Strong-AI-Lab/Logical-and-abstract-reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2310_09430
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Assessing and Enhancing the Robustness of Large Language Models with Task Structure Variations for Logical Reasoning
Bao, Qiming
Gendron, Gael
Peng, Alex Yuxuan
Zhong, Wanjun
Tan, Neset
Chen, Yang
Witbrock, Michael
Liu, Jiamou
Computation and Language
Artificial Intelligence
Large language models (LLMs), such as LLaMA, Alpaca, Vicuna, GPT-3.5 and GPT-4, have advanced the performance of AI systems on various natural language processing tasks to human-like levels. However, their generalisation and robustness when performing logical reasoning has not been sufficiently assessed. To comprehensively evaluate this ability, we develop three new logical reasoning datasets named "ReClor-plus", "LogiQA-plus" and "LogiQAv2-plus" that extend standard logical reasoning datasets to evaluate the robustness of the LLM's reasoning. For each, we create three subsets: the first with randomly shuffled options, the second with the correct choices replaced by "none of the other options is correct", and the third with a combination of shuffling and substitution. Experiments on these datasets show that these simple augmentations greatly hinder the models' performance. Despite their high performance on the original publicly available datasets, we find that all models perform poorly on these newly constructed datasets. We also demonstrate that introducing task variations into the training set can markedly improve the model's performance on both the original and our developed datasets. Finally, we show that applying logic-driven data augmentation for fine-tuning and prompting can enhance generalisation in both discriminative and generative models, offering a path to improving their robustness for tasks involving logical reasoning. Source code and data are made publicly available at https://github.com/Strong-AI-Lab/Logical-and-abstract-reasoning.
title Assessing and Enhancing the Robustness of Large Language Models with Task Structure Variations for Logical Reasoning
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2310.09430