Author:
Tang Zhisheng,Kejriwal Mayank
Abstract
AbstractWe present a detailed case study evaluating selective cognitive abilities (decision making and spatial reasoning) of two recently released generative transformer models, ChatGPT and DALL-E 2. Input prompts were constructed following neutral a priori guidelines, rather than adversarial intent. Post hoc qualitative analysis of the outputs shows that DALL-E 2 is able to generate at least one correct image for each spatial reasoning prompt, but most images generated are incorrect, even though the model seems to have a clear understanding of the objects mentioned in the prompt. Similarly, in evaluating ChatGPT on the rationality axioms developed under the classical Von Neumann-Morgenstern utility theorem, we find that, although it demonstrates some level of rational decision-making, many of its decisions violate at least one of the axioms even under reasonable constructions of preferences, bets, and decision-making prompts. ChatGPT’s outputs on such problems generally tended to be unpredictable: even as it made irrational decisions (or employed an incorrect reasoning process) for some simpler decision-making problems, it was able to draw correct conclusions for more complex bet structures. We briefly comment on the nuances and challenges involved in scaling up such a ‘cognitive’ evaluation or conducting it with a closed set of answer keys (‘ground truth’), given that these models are inherently generative and open-ended in responding to prompts.
Publisher
Springer Science and Business Media LLC
Reference32 articles.
1. Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN et al. Attention is all you need. Adv Neural Informat Process Syst. 2017; 30.
2. Devlin J, Chang MW, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. 2018; arXiv preprint arXiv:1810.04805.
3. Liu Y, Ott M, Goyal N, Du J, Joshi M, Chen D et al. Roberta: A robustly optimized bert pretraining approach. 2019; arXiv preprint arXiv:1907.11692.
4. Bianchi F, Kalluri P, Durmus E, Ladhak F, Cheng M, Nozza D et al. Easily accessible text-to-image generation amplifies demographic stereotypes at large scale. 2022; arXiv preprint arXiv:2211.03759.
5. Ettinger A. What BERT is not: lessons from a new suite of psycholinguistic diagnostics for language models. Trans Assoc Comput Linguist. 2020;8:34–48.