Ivan Titov

Professor in Natural Language Processing & ML · Edinburgh & Amsterdam

  • Sparsity and differentiable structure: sparse attention, binary variables, adaptive computation, neuron pruning
  • Knowledge in LLMs: editing without retraining, unlearning, memorisation, knowledge offloading
  • Interpretability and trust: differentiable masking, representation analysis, chain-of-thought faithfulness, controllability
  • Post-training methods: routing, modularity, reasoning
  • Communication between foundation models: learned latent interfaces, agent collaboration

Earlier interests include graph neural networks, structured prediction, latent variable models, multilinguality, machine translation, federated/privacy-preserving NLP, and semantic parsing.

🎓 Prospective students — I am recruiting

I am looking for PhD students — apply via ILCC, Centre for Doctoral Training (CDT) in Responsible NLP, or CDT in ML Systems. See the joining the lab page for more information, and the lab page for current students and where our alumni have gone next — faculty positions and labs such as Google DeepMind and Meta AI.

I direct the CDT in NLP at Edinburgh and the Edinburgh ELLIS unit; ELLIS Fellow, former Turing Fellow, co-director of the ELLIS NLP program. Program Co-Chair at ICLR 2021 and CoNLL 2018; Action Editor for JMLR and TACL; paper awards at ACL and EMNLP.

Funded by ERC, NWO (Vici), UKRI, Google, SAP, Booking.com, Amazon.

News