Is your counterparty a neural network?

Authors: Kuparenkov I.A.
Published in issue: #3(92)/2024
DOI:
Category: Informatics, Computer Engineering and Control | Chapter: Information Technology. Computer techologies. Theory of computers and systems
Keywords: transformer architecture, large language model, text generation, artificial intelligence, chatbot identification method, neural network, Turing test, chatbot
Published: 17.07.2024

Widespread dissemination of the chatbots based on the large language models capable of generating texts and maintaining a conversation caused the problem of identifying the counterparty. The paper examines architecture used in constructing modern chatbots and studies their capabilities and methods of application. Based on the knowledge obtained, the author suggests distinguishing a chatbot from a person by testing the counterparty for the ability to think based on the Turing test. The experiments performed proved that this method was demonstrating high practical efficiency.

References

[1] Vaswani A., Noam Shazeer, Niki Parmar et al. Attention is all you need. Advances in neural information processing systems. 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, 2017, vol. 30.

[2] Liu P.J. et al. Generating wikipedia by summarizing long sequences. arXiv preprint arXiv:1801.10198, 2018. https://doi.org/10.48550/arXiv.1801.10198

[3] Kitaev N., Klein D. Constituency parsing with a self-attentive encoder. arXiv preprint arXiv:1805.01052, 2018. https://doi.org/10.48550/arXiv.1805.01052

[4] Radford A., Karthik Narasimhan, Tim Salimans et al. Improving language understanding by generative pre-training. Available at: https://www.cs.ubc.ca/~amuham01/LING530/papers/radford2018improving.pdf (accessed April 15, 2024).

[5] Brown T., Mann B., Ryder N. et al. Language models are few-shot learners. Advances in neural information processing systems, 2020, vol. 33, pp. 1877–1901.

[6] Kozlova A.A., Popov I.A. Development of artificial intelligence technologies in Russia. Problems of modern society through the eyes of young researchers. XV All-Russian scientific-practical. conf. collection of materials. Volgograd, University book, 2023, pp. 341–343. (In Russ.).

[7] Goncharov D.S., Grigoriev S.V. Large language models on the example of chatbots GPT-3: today

[8] Turing A. Computing machinery and intelligence. Mind, 1950, no. 59, pp. 433–460.

[9] Log of testing the first hypothesis. URL: https://poe.com/s/61PCYXU7EzY9VZ2c0uhI (accessed 10.06.2023).

[10] Log of the second hypothesis verification. URL: https://poe.com/s/WCggrijHISPqw9hyPprB (accessed 10.06.2023).