"The future is already here — it's just not very evenly distributed"
- William Gibson -
I am interested in the safe, robust, and effective integration of machine learning systems into society. I primarily explore this topic through language models, causal representation learning, and adaptive computation. A recurring research question of interest is:
"How does machine intelligence differ from human intelligence?"
My past research has predominantly focused on unsupervised (deep) generative modeling with a special interest in methods exploring (non-trivial) manifold learning.
Reviewer for ICLR ’21, NeurIPS ’22, ICML ’23, ‘24 (best reviewer), and various workshops.
Selected Research
(see Google Scholar for an up-to-date list)
T.R. Davidson, V. Surkov, V. Veselovsky, G. Russo, R. West, C. Gulcehre,
Self-Recognition in Language Models, (EMNLP, 2024)
[arXiv] [code] [blog] [press]
T.R. Davidson*, V. Veselovsky*, M. Josifoski, M. Peyrard, A. Bosselut, M. Kosinski, R. West,
Evaluating Language Model Agency through Negotiations, (ICLR, 2024)
[arXiv] [code] [blog] [data]
T.R. Davidson*, L. Falorsi*, N. De Cao*, T. Kipf, J.M. Tomczak,
Hyperspherical Variational Autoencoders, Oral presentation (UAI, 2018)
[arXiv] [code] [oral]
L. Falorsi*, P. de Haan*, T.R. Davidson*, N. De Cao, M. Weiler, P. Forré, T. Cohen,
Explorations in Homeomorphic Variational Auto-Encoding, (ICML Workshop, 2018)
[arXiv] [code]
L. Falorsi, P. de Haan, T.R. Davidson, P. Forré,
Reparameterizing Distributions on Lie Groups, Oral presentation (AISTATS, 2019)
[arXiv] [code] [slides]