Datasets:

Modalities:
Tabular
Text
Languages:
English
ArXiv:
License:
OttoTarkka commited on
Commit
6cf8e88
·
verified ·
1 Parent(s): 6c8620f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +18 -8
README.md CHANGED
@@ -33,9 +33,11 @@ configs:
33
 
34
  ![Descriptor logo](DocDescriptorLogo_medium.png)
35
 
36
- Data release for the paper Task-Agnostic Web Document Annotation with LLM-Generated Descriptors (forthcoming).
 
 
 
37
 
38
- The descriptors are generated via a task-agnostic data annotation pipeline described in the paper (link coming soon).
39
 
40
  This Hugging Face dataset repository contains 5 distinct datasets:
41
  1. a descriptor-annotated version of a 10 billion token (~15 million document) [sample of FineWeb](https://huggingface.co/datasets/HuggingFaceFW/fineweb/viewer/sample-10BT).
@@ -45,15 +47,23 @@ This Hugging Face dataset repository contains 5 distinct datasets:
45
  5. Merge traces, i.e. the final schema with pointers to the original raw descriptors that lead to the final descriptor in the merging steps.
46
 
47
  ### Citing
48
- If you use this data please cite the original FineWeb paper and the our paper.
49
 
50
  **Task-Agnostic Web Document Annotation with LLM-Generated Descriptors**
51
  ```text
52
- @inproceedings{
53
- tarkkaetal-descriptors,
54
- title={Task-Agnostic Web Document Annotation with LLM-Generated Descriptors},
55
- author={Tarkka, Otto and Henriksson, Erik and Kanerva, Jenna and Ginter, Filip},
56
- year={2026 (forthcoming)},
 
 
 
 
 
 
 
 
57
  }
58
  ```
59
 
 
33
 
34
  ![Descriptor logo](DocDescriptorLogo_medium.png)
35
 
36
+ Data release for the paper Task-Agnostic Web Document Annotation with LLM-Generated Descriptors.
37
+
38
+ The descriptors are generated via a task-agnostic data annotation pipeline described in [the paper](https://aclanthology.org/2026.wac-1.24/).
39
+ The GitHub repo for this project can be found [here](https://github.com/TurkuNLP/LLM_document_descriptors/).
40
 
 
41
 
42
  This Hugging Face dataset repository contains 5 distinct datasets:
43
  1. a descriptor-annotated version of a 10 billion token (~15 million document) [sample of FineWeb](https://huggingface.co/datasets/HuggingFaceFW/fineweb/viewer/sample-10BT).
 
47
  5. Merge traces, i.e. the final schema with pointers to the original raw descriptors that lead to the final descriptor in the merging steps.
48
 
49
  ### Citing
50
+ If you use this data please use the following citations:
51
 
52
  **Task-Agnostic Web Document Annotation with LLM-Generated Descriptors**
53
  ```text
54
+ @inproceedings{tarkka-etal-2026-task,
55
+ title = "Task-Agnostic Web Document Annotation with {LLM}-Generated Descriptors",
56
+ author = "Tarkka, Otto and Henriksson, Erik and Kanerva, Jenna and Ginter, Filip",
57
+ editor = "Ortiz Suarez, Pedro and Burchell, Laurie and Ljube{\v{s}}i{\'c}, Nikola and Scherrer, Yves and Laippala, Veronika and Dini{\'c}, Vuk and Vaughan, Thom",
58
+ booktitle = "Proceedings of the 13th Web-as-Corpus Workshop",
59
+ month = oct,
60
+ year = "2026",
61
+ address = "Budapest, Hungary",
62
+ publisher = "Association for Computational Linguistics",
63
+ url = "https://aclanthology.org/2026.wac-1.24/",
64
+ doi = "10.18653/2026.wac-1.24",
65
+ pages = "252--281",
66
+ ISBN = "979-8-89176-481-1",
67
  }
68
  ```
69