The Harry Potter books are proving to be a rich resource for experimenting with language models due to their enduring influence in popular culture and the intricate wordplay found within their pages.
One notable study, titled “Who’s Harry Potter?,” explores a groundbreaking technique that allows large language models to selectively forget information. Microsoft researchers Mark Russinovich and Ronen Eldan demonstrated that AI models can be modified to erase knowledge of the Harry Potter books—characters, plots, and all—while maintaining their overall decision-making and analytical abilities. The choice of Harry Potter was deliberate, as the series is universally familiar, making it easy for researchers to evaluate the effectiveness of the technique.
The AI industry faces challenges regarding copyrighted material and problematic content in the vast datasets used to train large language models, which power AI chatbots. This has led to legal issues and public scrutiny for some AI companies. The research by Russinovich and Eldan suggests a potential solution by showing that AI models can unlearn specific content without compromising their functionality.
Another study conducted by researchers from the University of Washington, University of California at Berkeley, and the Allen Institute for AI introduces a language model named Silo. This model aims to reduce legal risks associated with data by selectively removing information. However, the researchers found that Silo’s performance suffered when trained solely on low-risk text. To investigate further, they turned to the Harry Potter books to analyze how individual pieces of text impact an AI system’s performance.
The researchers created two datastores—one including all published books except the first Harry Potter book, and another including all books in the series except the second, and so on. Removing the Harry Potter books from the datastore resulted in a decline in the model’s accuracy, as measured by perplexity, a metric used to evaluate AI models.
The use of Harry Potter in AI studies has been present for at least a decade, but it has become increasingly common as researchers focus on developing AI tools that can understand and respond to natural language effectively. The richness of scenes, dialogs, and emotional moments in the Harry Potter series makes it particularly relevant to the field of natural language processing.
Recent papers on arXiv, an open-access repository of scientific research, include intriguing titles such as “Machine Learning for Potion Development at Hogwarts,” “Large Language Models Meet Harry Potter,” and “Detecting Spells in Fantasy Literature with a Transformer-Based Artificial Intelligence.”
Even when not at the center of the research, Harry Potter remains a favorite literary reference for researchers. In one study, Rowling’s works were used to test the intelligence of AI systems, including those behind the creation of ChatGPT—a chatbot that has sparked lively debates. Terrence Sejnowski, from the Salk Institute for Biological Studies, likened chatbots to the Mirror of Erised in the first Harry Potter book, suggesting that they reflect the intelligence and biases of their users.
(With inputs from Bloomberg)
Unlock a world of Benefits! From insightful newsletters to real-time stock tracking, breaking news and a personalized newsfeed – it’s all here, just a click away! Login Now!
Published: 27 Dec 2023, 10:58 AM IST
/cdn.vox-cdn.com/uploads/chorus_asset/file/25263501/STK_414_AI_A.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/25439282/640.jpg)
