Teaching
VI Tutorial Deep Learning for NLP
News
- 2024
- Two papers accepted at EACL2024! Come talk to us about uncertainty in NLP, we will be presenting about methodology to evaluate LM’s calibration against human uncertainty and confidence vs. label variation, see you in Malta!
- UncertaiNLP’s programme has been finalised, we have a great line-up of invited talks and accepted papers. See you in Malta!
- 2023
- The first edition of our workshop on uncertainty-aware NLP is going to be in Malta! Check our cfp!
- The work we started at the Third ELLIS NLP Workshop in Dagstuhl is now out, check our pre-print!
- Evaluating uncertainty in neural text generators has been accepted at EMNLP2023!
- Vision DiffMask (faithful attributions for visual transformers) will appear at XAI4CV, go talk to Tinos at CVPR!
- My daughter was born <3 (I guess this space is about to get a bit quieter, unlike my home :p)!
- 2022
- Two papers accepted at EMNLP 2022: sampling-based MBR for NMT and calibration of text classifiers under human disagreement
I am hiring a PhD candidate to work on uncertainty and language generation, apply before 21 August 2022!- I gave a talk at Structure Prediction for NLP (slides, colab demo).
- Deep Learning 2 course is ready!
- Sparse communication via mixed distributions has been accepted at ICLR2022!
- 2021
- 2 EMNLP2021 papers led by Nicola De Cao: Editing Factual Knowledge in Language Models and Highly Parallel Autoregressive Entity Linking with Discriminative Correction.
- In this followup to our Coling2020 paper we take MBR decoding from proof-of-concept to a robust and scalable decision rule.
- Another collaboration with DeepSPIN! This time we provide a formal account to mixed random variables and take the first steps towards a theory of communication using sparse continuous random variables.
PhD and postdoc positions at the newly established Mercury Machine Learning Lab (a collaboration among Booking.com, TUDelft, and UvA) including 2 PhD position in ML/NLP, together with Ivan Titov. Deadline: August 23, 2021.- I gave a talk at the IST and Unbabel Seminars
- I gave a talk at JHU’s Center for Language and Speech Processing
- I gave a talk at the University of Helsinki’s Language Technology Seminar
- 2020
- The inadequacy of the mode in machine translation (Coling2020, best paper).
- Exact gradients via sparse marginalisation (NeurIPS2020, spotlight), a collaboration with DeepSPIN.
- Learn about attribution in deep NNs with differentiable masking (EMNLP2020).
- A new distribution on hyperspheres (INNF+2020).
- Check our work on estimation of deep generative LMs (ACL2020).
- Our latent morphology model for translation got a spotlight at ICLR2020.