Conveners
Text+ Plenary Tag 1: Nachmittagssession
- Andreas Henrich
Der Vortrag widmet sich dem - insbesondere auch nichtwissenschaftlichen - Impact wissenschaftlicher Forschung und thematisiert inwieweit dieser klassifiziert, analysiert und automatisiert vorhergesagt werden kann.
Während die Erfassung wissenschaftlichen Impacts bisher hauptsächlich auf der Analyse wissenschaftlicher Veröffentlichungen und ihrer Verbreitung beruhte, verwendet der...
The talk discusses prompting as a method to use large language models for applications in the computational literary studies. With examples and experimental results taken from the Q:TRACK project, it will explain assumptions and possibilities, but also pitfalls and challenges. The talk closes with a number of recommendations for users.
Recently, language technology has seen tremendous advancements due to the development and use of large language models, large machine learning models pre-trained on large amounts of textual data. However, while how humans express themselves in language and how they perceive language is largely driven by their individual sociodemographic and sociocultural backgrounds, language models still only...
A brief into on how Sparse Autoencoders (SAE) can be leveraged to extract interpretable, monosemantic features from the opaque intermediate activations of LLMs, providing a window into their internal representations. And we hope to initiate discussions on the methodology of training SAEs on LLM activations, the resulting sparse and high-dimensional representations, and how these can be...
