| 2026 | WWW | Tuning Qwen2.5-VL to Improve Its Web Interaction Skills. | Alexandra Yakovleva, Henrik Prssinen, Harri Valpola, Juho Kannala, Alexander Ilin |
| 2017 | ICLR | Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. | Antti Tarvainen, Harri Valpola |
| 2013 | ICONIP | Pushing Stochastic Gradient towards Second-Order Methods - Backpropagation Learning with Transformations in Nonlinearities. | Tommi Vatanen, Tapani Raiko, Harri Valpola, Yann LeCun |
| 2009 | ICANN | Selective Attention Improves Learning. | Antti Yli-Krekola, Jaakko Srel, Harri Valpola |
| 2006 | IJCNN | Extraction of Components with Structured Variance. | Alexander Ilin, Harri Valpola, Erkki Oja |
| 2005 | IJCNN | Semiblind source separation of climate data detects El Nino as the component with the highest interannual variability. | Alexander Ilin, Harri Valpola, Erkki Oja |
| 2005 | UAI | Bayes Blocks: An Implementation of the Variational Bayesian Building Blocks Framework. | Markus Harva, Tapani Raiko, Antti Honkela, Harri Valpola, Juha Karhunen |
| 2001 | ICANN | Independent Variable Group Analysis. | Krista Lagus, Esa Alhoniemi, Harri Valpola |