Hi, I don't fully understand the concern. The two tracks are entirely separate; any overlap in texts is coincidental. Feel free to reach out to us via email with more details: narrative-similarity-task-organizers@googlegroups.com
We will release it today, sorry!
There is still time to get started on the shared task. The test phase will be held in January.
This week you can find many of us organizers at @comphumresearch.bsky.social. Feel free to reach out; look for @evelyngius.bsky.social, @haimostiemer.bsky.social, and @hanshatzel.bsky.social.
Now on our website: synthetic training data! We expect this to be especially useful for Track B, but we are eager to see what you do with it!
You are, of course, also allowed and encouraged to create your own synthetic training data.
Our development data is now available, and the competition has opened on CodaBench. Whether it's chat models, symbolic story representations, or something else entirely, every approach is welcome. We are looking forward to your submissions!
www.codabench.org/competitions...
🚨 Shared Task Alert! 🚨
We are announcing the shared task on narrative similarity: SemEval-2026 Task 4: Narrative Story Similarity and Narrative Representation Learning
We invite you to benchmark LLMs, embedding models, or even test your favorite narrative formalism. Sample data is now available!
