565 citations · 1k across the 23 of their papers we have counts for
5 papers · 1 filter
Sampling from Discrete Energy-Based Models with Quality/Efficiency Trade-offs
Bryan Eikema, Germán Kruszewski, Hady Elsahar +1
Energy-Based Models (EBMs) allow for extremely flexible specifications of probability distributions. However, they do not provide a mechanism for obtaining exact samples from these…
Controlling Conditional Language Models without Catastrophic Forgetting
Tomasz Korbak, Hady Elsahar, German Kruszewski +1
Machine learning is shifting towards general-purpose pretrained generative models, trained in a self-supervised manner on large amounts of data, which can then be applied to solve…
Unsupervised and Distributional Detection of Machine-Generated Text
Matthias Gallé, Jos Rozen, Germán Kruszewski +1
The power of natural language generation models has provoked a flurry of interest in automatic methods to detect if a piece of text is human or machine-authored. The problem so far…
Energy-Based Models for Code Generation under Compilability Constraints
Tomasz Korbak, Hady Elsahar, Marc Dymetman +1
Neural language models can be successfully trained on source code, leading to applications such as code completion. However, their versatile autoregressive self-supervision objecti…
Emergence of Self-Reproducing Metabolisms as Recursive Algorithms in an Artificial Chemistry
Germán Kruszewski, Tomas Mikolov
One of the main goals of Artificial Life is to research the conditions for the emergence of life, not necessarily as it is, but as it could be. Artificial Chemistries are one of th…