I'm a postdoc working with Dr. Ivan Titov.

I'm interested in neural networks and deep learning. My current work is to employ them to solve natural language processing tasks such as coreference resolution, dependency parsing, semantic role labelling, and sentiment analysis.  I'm also interested in formal semantics, especially learning semantic parsing, which is the topic of my master thesis.
I did a PhD at the same university, under supervision of Dr. (Jelle) Willem Zuidema. My promotor is Prof. Rens Bod.

Checkout my CV here and Google Scholar


Natural Language Processing

  • P. Le, I. Titov. Optimizing Differentiable Relaxations of Coreference Evaluation Metrics. CoNLL (to appear). [pdf, bib, code]
  • P. Le, M. Dymetman, J-M. Renders. LSTM-based Mixture-of-Experts for Knowledge-Aware Dialogues. ACL Workshop on Representation Learning for NLP, 2016. [pdf, bib].
  • P. Le and W. Zuidema. Quantifying the vanishing gradient and long distance dependency problem in recursive neural networks and recursive LSTMs. ACL Workshop on Representation Learning for NLP, 2016. [pdf, bib]
  • P. Le and W. Zuidema. The Forest Convolutional Network: Compositional Distributional Semantics with a Neural Chart and without Binarization. EMNLP [pdfbib]
  • P. Le. Enhancing the Inside-Outside Recursive Neural Network Reranker for Dependency Parsing. International Conference on Parsing Technology (IWPT) [pdfbib]
  • P. Le and W. Zuidema. Compositional Distributional Semantics with Long Short Term MemoryJoint Conference on Lexical and Computational Semantics (*SEM) [pdfbibcode]
  • P. Le and W. Zuidema. Unsupervised Dependency Parsing: Let's Use Supervised Parsers. NAACL-HLT [pdfbib]


  • P. Le and W. Zuidema. Inside-Outside Semantics: A Framework for Neural Models of Semantic Composition. NIPS 2014 Workshop on Deep Learning and Representation Learning [pdf]
  • P. Le and W. Zuidema. The Inside-Outside Recursive Neural Network model for Dependency Parsing. EMNLP [pdfbibcode]


  • P. Le, W. Zuidema, and R.J.H. Scha. Learning from errors: Using vector-based compositional semantics for parse reranking. ACL Workshop on Continuous Vector Space Models and their Compositionality (oral) [pdfbib]


  • P. Le and W. Zuidema. Learning Compositional Semantics for Open Domain Semantic Parsing. COLING [pdfbib]

Computer Vision
  • P. Le, A.D. Duong, H.Q. Vu, and N.T. Pham (2009). Adaptive hybrid mean shift and particle filter. In proceedings of International Conference on Computing and Communication Technologies, 2009. RIVF’09, pages 1–4. [pdf]
  • P. Le,  N.T. Pham, A.D. Duong, and H.Q. Vu (2008). Tracking a variable number of humans and motorcycles in video via Probability Hypothesis Density filter. IEEE Student Paper Contest. Seoul, Korea.

  • P. Le. Learning Vector Representations for Sentences - The Recursive Deep Learning Approach. PhD's Thesis, University of Amsterdam[pdf]
  • P. Le. Learning Semantic Parsing. Master's Thesis, University of Amsterdam. [pdf] (STIL Thesis Prize)