17 februari 2024. the case for a small language network is de nieuwe installatie die beeldend kunstenaar, ontwerper en programmeur Richard Vijgen maakte op verzoek van en in samenwerking met Rozalie Hirs op basis van haar dichtbundel oneindige zin. De installatie toont talloze leesmogelijkheden van de bundel. De programmatuur van de hand van Richard Vijgen maakt dat de installatie gaandeweg leert via Artificial Intelligence. Een eerder prototype voor de installatie is tot stand gekomen met financiële ondersteuning door het Nederlands Letterenfonds in het kader van Digitale literaire projecten.

The Case For a Small Language Network (2024)
Richard Vijgen – installation, concept
Rozalie Hirs – poetry, concept
Jelle Reith – technical realisation
Supported by the Netherlands Foundation for Literature & the Creative Industries Fund Netherlands

When a language model produces a sentence it presents us a statistical probability based on countless texts it has analysed. Before it is able to predict the next character in a sentence it has to cut the writings of millions of authors into fragments to analyse the sequence of characters. They are stripped of meaning and structure and repurposed as a statistical resource. While the system needs the author’s work as input, the results can never be traced back. But what would happen when a language model creates new texts while leaving the original work intact? What if generative AI can be traced back to and understood in the context of the original text?

The Case For a Small Language Network is a speculative AI, based on the work of Dutch composer and poet Rozalie Hirs. Her 2021 poetry book oneindige zin (which can be read as infinite sense or infinite phrase in Dutch) can be read as one never ending phrase. The installation shows the entire book printed on five 30 meter long strips of labelprinter paper that scroll in both directions. As the five lines move back and forth, a vertical reading allows for new combinations to emerge.

Meanwhile a neural network based on Andrej Karpathy’s Char-RNN analyses a digital copy of Hirs’ original text and tries to create new sentences based on her work. Initially the combinations seem random and nonsensical but as the training of the neural network (running on a low power Raspberry Pi) progresses, more interesting combinations emerge. Rather than appropriating the the authors work as mere statistical data and cutting it into fragments, the system leaves the original text intact. It’s output can only be read and understood in the context of the input, as the only way to display it is to move the entire manuscript text left or right.

The Case for a Small Language Network reflects on the role of authorship in generative AI and questions the practice of reducing the written expressions of millions of authors, (mostly without their permission) into a statistical resource.