Can language representation models think in bets?

Author:

Tang Zhisheng1,Kejriwal Mayank1ORCID

Affiliation:

1. Information Sciences Institute, USC Viterbi School of Engineering, 4676 Admiralty Way 1001, Marina Del Rey, CA 90292 USA

Abstract

In recent years, transformer-based language representation models (LRMs) have achieved state-of-the-art results on difficult natural language understanding problems, such as question answering and text summarization. As these models are integrated into real-world applications, evaluating their ability to make rational decisions is an important research agenda, with practical ramifications. This article investigates LRMs’ rational decision-making ability through a carefully designed set of decision-making benchmarks and experiments. Inspired by classic work in cognitive science, we model the decision-making problem as a bet. We then investigate an LRM’s ability to choose outcomes that have optimal, or at minimum, positive expected gain. Through a robust body of experiments on four established LRMs, we show that a model is able to ‘think in bets’ if it is first fine-tuned on bet questions with an identical structure. Modifying the bet question’s structure, while still retaining its fundamental characteristics, decreases an LRM’s performance by more than 25%, on average, although absolute performance remains well above random. LRMs are also found to be more rational when selecting outcomes with non-negative expected gain, rather than optimal or strictly positive expected gain. Our results suggest that LRMs could potentially be applied to tasks that rely on cognitive decision-making skills, but that more research is necessary before these models can robustly make rational decisions.

Funder

Defense Advanced Research Projects Agency

Publisher

The Royal Society

Subject

Multidisciplinary

Reference107 articles.

1. Devlin J Chang MW Lee K Toutanova K. 2018 BERT: pre-training of deep bidirectional transformers for language understanding. (https://arxiv.org/abs/1810.04805)

2. Language models are unsupervised multitask learners;Radford A;OpenAI blog,2019

3. Language models are few-shot learners;Brown T;Adv. Neural Inf. Process. Syst.,2020

4. MatSciBERT: A materials domain language model for text mining and information extraction

5. Larionov D Shelmanov A Chistova E Smirnov I. 2019 Semantic role labeling with pretrained language models for known and unknown predicates. In Proc. of the Int. Conf. on Recent Advances in Natural Language Processing (RANLP 2019) Varna Bulgaria September pp. 619–628. INCOMA Ltd.

Cited by 4 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3