Add README
Browse files
README.md
CHANGED
|
@@ -1,3 +1,31 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
# StoryEmb
|
| 6 |
+
|
| 7 |
+
This is the model from [our paper](https://www.inf.uni-hamburg.de/en/inst/ab/lt/publications/2024-hatzel-et-al-emnlp.pdf) "Story Embeddings — Narrative-Focused Representations of Fictional Stories". We publish the variant trained on augmented data.
|
| 8 |
+
|
| 9 |
+
You can use this just like the base model [intfloat/e5-mistral-7b-instruct](https://huggingface.co/intfloat/e5-mistral-7b-instruct).
|
| 10 |
+
We trained with the task prefix "Retrieve stories with a similar narrative to the given story: " so we recommand that you use this for story retrieval.
|
| 11 |
+
|
| 12 |
+
If you are interested in just the adapter weights (e.g. for continued fine-tuning), check out the directory `adapter-weights`.
|
| 13 |
+
|
| 14 |
+
Do not hestitate to reach out if you encounter any issues! You are likely one of the first few people actually downloading this :)
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
## Citation
|
| 18 |
+
|
| 19 |
+
If you are making use of this model in your publication, please cite our paper:
|
| 20 |
+
|
| 21 |
+
```
|
| 22 |
+
@inproceedings{hatzel-biemann-2024-story-embeddings,
|
| 23 |
+
title = "Story Embeddings -- Narrative-Focused Representations of Fictional Stories",
|
| 24 |
+
author = "Hatzel, Hans Ole and Biemann, Chris",
|
| 25 |
+
booktitle = "Proceedings of the 62st Annual Meeting of the Association for Computational Linguistics",
|
| 26 |
+
year = "2024",
|
| 27 |
+
address = "Miami, Florida",
|
| 28 |
+
publisher = "Association for Computational Linguistics",
|
| 29 |
+
}
|
| 30 |
+
```
|
| 31 |
+
|