# A Pilot Evaluation of ChatGPT and DALL-E 2 on Decision Making and Spatial Reasoning

arXiv: https://arxiv.org/abs/2302.09068 · HTML: https://mlsanity.com/paper/2302.09068/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 15 of its 27 references that are in the map.
- First posted on arXiv 15 Feb 2023.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)

## Builds on (most cited first, 15 of 15)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://mlsanity.com/paper/2302.04023/) (2023)
- [A Primer in BERTology: What we know about how BERT works](https://mlsanity.com/paper/2002.12327/) (2020)
- [What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language models](https://mlsanity.com/paper/1907.13528/) (2019)
- [Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale](https://mlsanity.com/paper/2211.03759/) (2022)
- [A very preliminary analysis of DALL-E 2](https://mlsanity.com/paper/2204.13807/) (2022)
- [Adv-BERT: BERT is not robust on misspellings! Generating nature adversarial samples on BERT](https://mlsanity.com/paper/2003.04985/) (2020)
- [Structured, flexible, and robust: benchmarking and improving large language models towards more human-like behavior in out-of-distribution reasoning tasks](https://mlsanity.com/paper/2205.05718/) (2022)
- [Generated Faces in the Wild: Quantitative Comparison of Stable Diffusion, Midjourney and DALL-E 2](https://mlsanity.com/paper/2210.00586/) (2022)
- [DALL-E 2 Fails to Reliably Capture Common Syntactic Processes](https://mlsanity.com/paper/2210.12889/) (2022)
- [A Survey on Neural Network Language Models](https://mlsanity.com/paper/1906.03591/) (2019)
- [Can Language Representation Models Think in Bets?](https://mlsanity.com/paper/2210.07519/) (2022)
- [BERT Has Uncommon Sense: Similarity Ranking for Word Sense BERTology](https://mlsanity.com/paper/2109.09780/) (2021)

## Built on by (most cited first, 1 of 1)

- [Summary of ChatGPT-Related Research and Perspective Towards the Future of Large Language Models](https://mlsanity.com/paper/2304.01852/) (2023)

## Abstract (arXiv)

We conduct a pilot study selectively evaluating the 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.