Chatgpt: Jack Of All Trades, Master Of None

Kocoń Jan, Cichecki Igor, Kaszyca Oliwier, Kochanek Mateusz, Szydło Dominika, Baran Joanna, Bielaniewicz Julita, Gruza Marcin, Janz Arkadiusz, Kanclerz Kamil, Kocoń Anna, Koptyra Bartłomiej, Mieleszczenko-kowszewicz Wiktoria, Miłkowski Piotr, Oleksy Marcin, Piasecki Maciej, Radliński Łukasz, Wojtasik Konrad, Woźniak Stanisław, Kazienko Przemysław. Information Fusion 2023

[Paper]    
Applications Ethics And Bias Few Shot GPT Model Architecture Pretraining Methods Prompting RAG Transformer

OpenAI has released the Chat Generative Pre-trained Transformer (ChatGPT) and revolutionized the approach in artificial intelligence to human-model interaction. Several publications on ChatGPT evaluation test its effectiveness on well-known natural language processing (NLP) tasks. However, the existing studies are mostly non-automated and tested on a very limited scale. In this work, we examined ChatGPT’s capabilities on 25 diverse analytical NLP tasks, most of them subjective even to humans, such as sentiment analysis, emotion recognition, offensiveness, and stance detection. In contrast, the other tasks require more objective reasoning like word sense disambiguation, linguistic acceptability, and question answering. We also evaluated GPT-4 model on five selected subsets of NLP tasks. We automated ChatGPT and GPT-4 prompting process and analyzed more than 49k responses. Our comparison of its results with available State-of-the-Art (SOTA) solutions showed that the average loss in quality of the ChatGPT model was about 25% for zero-shot and few-shot evaluation. For GPT-4 model, a loss for semantic tasks is significantly lower than for ChatGPT. We showed that the more difficult the task (lower SOTA performance), the higher the ChatGPT loss. It especially refers to pragmatic NLP problems like emotion recognition. We also tested the ability to personalize ChatGPT responses for selected subjective tasks via Random Contextual Few-Shot Personalization, and we obtained significantly better user-based predictions. Additional qualitative analysis revealed a ChatGPT bias, most likely due to the rules imposed on human trainers by OpenAI. Our results provide the basis for a fundamental discussion of whether the high quality of recent predictive NLP models can indicate a tool’s usefulness to society and how the learning and validation procedures for such systems should be established.

Similar Work