Messages
The WNNLP'26 proceedings are now published, please have a look at the PDF.
The workshop received 29 submissions by 44 authors. We are glad that all of them have been accepted for publication as part of the WNNLP 2026 proceedings. This is the largest number of submissions throughout the long history of the WNNLP workshop. Slides for all the presentations.
The Programme Committee has selected the paper "Steering LLMs’ Responses Towards Moral Foundations on the Norwegian MFQ-30" by Hans Andersen and David Dichas for the WNNLP 2026 Best Paper award. In this paper, the authors conduct a thorough analysis of the behavior of six large language models on the Norwegian MFQ-30 morality questionnaire, with interesting ...
All WNNLP'26 teams with the exam papers accepted should prepare a "lightning talk" for the final workshop on June 1st at 10:15, including a few slides. PhD students will have 4 minutes each to present their paper (4 slides max); MSc teams will have 2 minutes each (3 slides max).
The track chairs will give a short task overview for each track, so you don't need to spend time and space on this in your presentation. Focus on the main contributions and results. Note that the time limit is strict due to our tight program, so be sure to not include more material than you can realistically cover.
The presentation slides should be pushed to the separate public (UiO-)git repository for the workshop. Make sure your slide...
The decisions for all the IN5550 exam paper submissions have now been made. Overall, we are impressed with the quality of your papers and are excited about the forthcoming WNNLP'26 workshop!
If your submission is "Accepted" or "Accepted conditionally", you should now upload the so-called "camera-ready" or final version of your paper. This is done via OpenReview by clicking on the Edit - Camera ready button. The possibility to submit the camera-ready version opens in the morning of May 28, and the deadline for this is May 28 anywhere on Earth (which means May 29, around noon Oslo time).
Before submitting the final version of your paper, you should do the following:
- Make sure the paper uses the ARR LaTeX te...
Hi all,
Here we publish the leaderboard for the second obligatory assignment. The 3 teams of Master students with the best performance scores will get one bonus point, if not already at the maximum.
Congratulations!
| Team | Krippendorff's alpha | Accuracy |
|---|---|---|
| Helene Brodin | 0.654 | 0.680 |
| Arthur Spalanzani | 0.643 | 0.664 |
| Muhammad Zeeshan Shahid, Ola Gabriel Huus Skeie, Sigurd Norbye | 0.632 | 0.736 |
The third obligatory assignment is now available in our GitHub repository. It deals with fine-tuning large generative language models to follow instructions.
The assignment is due April 17.
As usual, we will evaluate your submitted models on a held-out test set. After grading is done, we will publish a leaderboard of the best 3 teams in terms of their models' performance. These 3 teams' members will get one extra point to their grades (unless they are already at the maximum points for this assignment).
Good luck!
Hi all, here's the leaderboard for the first obligatory assignment. The 3 teams of Master students with the best performance scores will get one bonus point, if not already at the maximum.
Congratulations!
| Team | Spearman | MSE |
|---|---|---|
| Elias Trana, Ola Lømo Ellingsen | 0.665 | 0.005 |
| Hugo Hilde | 0.617 | 0.007 |
| Arthur Spalanzani | 0.602 | 0.007 |
The second obligatory assignment is now available in our GitHub repository. This assignment deals with the Word-in-Context (WiC) task: given a target word and two sentences containing it, determine how similar the word's meaning is in the two contexts.
The assignment is due March 20. It is mostly targeted at finetuning masked language models, which will be covered in the group session on March 4 (tomorrow).
Note that we will evaluate your submitted models on a held-out test set. After grading is done, we will publish a leaderboard of the best 3 teams in terms of their models' performance. These 3 teams' members will get one extra point to their grades (unless they are already at the maximum points for this assignment).
Good luck!
The first obligatory assignment is now online, on the Git repository. Your task is to build a neural predictor of emotional valence of English texts. This is a regression problem which you will solve using bags-of-words and word embeddings as features.
The assignment is due February 27, but feel free to start working on it. We have already covered linear models, feed-forward neural networks, basics of PyTorch and word embeddings in the lectures. We will also provide extensive practice in building neural classifiers with PyTorch and word embeddings at the next group sessions.
Note that we will evaluate your submitted models on a held-out test set. After grading is done, we will publish a leaderboard of the best 3 teams in terms of their models' performance. These 3 teams' members will get one extra point to their grades (unless they are already at the maximum of 6 points for ...
The slides for the introductory lecture we had on Monday are now published.
You can find them at the course web page under the schedule ("Schedule in TP"). Just find the lecture you need, click on it, and then scroll to "Syllabus".
On Wednesday (tomorrow) we will have the first group session. It starts at 10:15, OJD Datastue Fortress. We will focus on working with Fox, setting up our programming environment and submitting simple jobs.
Please make sure you have access to the course GitHub repository, and that you have applied to the Educloud project `ec403`, as described here. Without an Educloud account you will not be able to access Fox.
See you tomorrow!
Welcome to our IN5550/IN9550 course which will guide you through deep learning applications in natural language processing!
The first introductory lecture this term will be held on Monday, January 19, at 10:15 (OJD, Seminarrom Pascal). We will go through course logistics (including routines for assignments and the final project-based exam) and motivate the now dominant use of neural machine learning architectures in Natural Language Processing (and most other sub-fields of Artificial Intelligence). The first lecture will be in-person.
Subsequent lectures (after the introductory one) will be provided to you in a pre-recorded format. Currently, the plan is that the lecture videos will be published every Wednesday in the second half of the day. You can watch them whenever it is more convenient to you. Every Monday (at the official de...