1 code implementation • 1 Jun 2023 • Paul Soulos, Edward Hu, Kate McCurdy, Yunmo Chen, Roland Fernandez, Paul Smolensky, Jianfeng Gao
To facilitate the learning of these symbolic sequences, we introduce a differentiable tree interpreter that compiles high-level symbolic tree operations into subsymbolic matrix operations on tensors.
no code implementations • MTSummit 2021 • Paul Soulos, Sudha Rao, Caitlin Smith, Eric Rosen, Asli Celikyilmaz, R. Thomas McCoy, Yichen Jiang, Coleman Haley, Roland Fernandez, Hamid Palangi, Jianfeng Gao, Paul Smolensky
Machine translation has seen rapid progress with the advent of Transformer-based models.
1 code implementation • NAACL 2021 • Yichen Jiang, Asli Celikyilmaz, Paul Smolensky, Paul Soulos, Sudha Rao, Hamid Palangi, Roland Fernandez, Caitlin Smith, Mohit Bansal, Jianfeng Gao
On several syntactic and semantic probing tasks, we demonstrate the emergent structural information in the role vectors and improved syntactic interpretability in the TPR layer outputs.
no code implementations • NeurIPS Workshop SVRHM 2020 • Paul Soulos, Leyla Isik
We further find that the latent dimensions in these models map onto non-overlapping regions in fMRI data, allowing us to "disentangle" different features such as 3D rotation, skin tone, and facial expression in the human brain.
2 code implementations • EMNLP (BlackboxNLP) 2020 • Paul Soulos, Tom McCoy, Tal Linzen, Paul Smolensky
How can neural networks perform so well on compositional tasks even though they lack explicit compositional representations?
no code implementations • 19 May 2018 • Joshua C. Peterson, Paul Soulos, Aida Nematzadeh, Thomas L. Griffiths
Modern convolutional neural networks (CNNs) are able to achieve human-level object classification accuracy on specific tasks, and currently outperform competing models in explaining complex human visual representations.