Hugging Face's logo

Datasets:
viniferagy
/
MoNIM

Modalities:
Text
Formats:
text
Languages:
English
Size:
1M - 10M
Tags:
continual-learning
language-modeling
knn-lm
semiparametric-models
news-crawl
monim
Libraries:
Datasets
License:
Dataset card Data Studio Files
xet
 Community
1
viniferagy commited on
Commit
397a581
·
1 Parent(s): fa77c41

Upload MoNIM days 1-30 dataset

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +5 -0
  2. README.md +79 -0
  3. daily/1/bin/dict.txt +3 -0
  4. daily/1/bin/preprocess.log +3 -0
  5. daily/1/bin/test.bin +3 -0
  6. daily/1/bin/test.idx +3 -0
  7. daily/1/bin/train.bin +3 -0
  8. daily/1/bin/train.idx +3 -0
  9. daily/1/bin/valid.bin +3 -0
  10. daily/1/bin/valid.idx +3 -0
  11. daily/1/fertility_cache_id.pickle +3 -0
  12. daily/1/freq_cache_id.pickle +3 -0
  13. daily/1/net/test.bpe +3 -0
  14. daily/1/net/test.txt +3 -0
  15. daily/1/net/total/bin/dict.txt +3 -0
  16. daily/1/net/total/bin/preprocess.log +3 -0
  17. daily/1/net/total/bin/test.bin +3 -0
  18. daily/1/net/total/bin/test.idx +3 -0
  19. daily/1/net/total/bin/train.bin +3 -0
  20. daily/1/net/total/bin/train.idx +3 -0
  21. daily/1/net/total/test.bpe +3 -0
  22. daily/1/net/total/train.bpe +3 -0
  23. daily/1/net/train.bpe +3 -0
  24. daily/1/net/train.txt +3 -0
  25. daily/1/test.bpe +3 -0
  26. daily/1/test.txt +3 -0
  27. daily/1/train.bpe +3 -0
  28. daily/1/train.txt +3 -0
  29. daily/1/valid.bpe +3 -0
  30. daily/1/valid.txt +3 -0
  31. daily/10/bin/dict.txt +3 -0
  32. daily/10/bin/preprocess.log +3 -0
  33. daily/10/bin/test.bin +3 -0
  34. daily/10/bin/test.idx +3 -0
  35. daily/10/bin/train.bin +3 -0
  36. daily/10/bin/train.idx +3 -0
  37. daily/10/bin/valid.bin +3 -0
  38. daily/10/bin/valid.idx +3 -0
  39. daily/10/fertility_cache_id.pickle +3 -0
  40. daily/10/freq_cache_id.pickle +3 -0
  41. daily/10/net/test.bpe +3 -0
  42. daily/10/net/test.txt +3 -0
  43. daily/10/net/total/bin/dict.txt +3 -0
  44. daily/10/net/total/bin/preprocess.log +3 -0
  45. daily/10/net/total/bin/test.bin +3 -0
  46. daily/10/net/total/bin/test.idx +3 -0
  47. daily/10/net/total/bin/train.bin +3 -0
  48. daily/10/net/total/bin/train.idx +3 -0
  49. daily/10/net/total/test.bpe +3 -0
  50. daily/10/net/total/train.bpe +3 -0
.gitattributes CHANGED
@@ -58,3 +58,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
58
  # Video files - compressed
59
  *.mp4 filter=lfs diff=lfs merge=lfs -text
60
  *.webm filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
58
  # Video files - compressed
59
  *.mp4 filter=lfs diff=lfs merge=lfs -text
60
  *.webm filter=lfs diff=lfs merge=lfs -text
61
+ *.txt filter=lfs diff=lfs merge=lfs -text
62
+ *.bpe filter=lfs diff=lfs merge=lfs -text
63
+ *.idx filter=lfs diff=lfs merge=lfs -text
64
+ *.json filter=lfs diff=lfs merge=lfs -text
65
+ *.log filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ language:
4
+ - en
5
+ pretty_name: MoNIM
6
+ tags:
7
+ - continual-learning
8
+ - language-modeling
9
+ - knn-lm
10
+ - semiparametric-models
11
+ - news-crawl
12
+ - monim
13
+ - memory
14
+ ---
15
+
16
+ # MoNIM
17
+
18
+ This dataset repository contains days 1-30 of the daily data used for reproducing the continual-learning experiments in the ACL 2025 paper [Learn to Memorize: Scalable Continual Learning in Semiparametric Models with Mixture-of-Neighbors Induction Memory](https://aclanthology.org/2025.acl-long.1385/).
19
+
20
+ ## Dataset Summary
21
+
22
+ The paper uses News Crawl data from the first half of 2020 (NC-20H1) as streaming data for continual learning. This repository provides the first 30 daily shards from the prepared `daily` data, intended for reproduction research.
23
+
24
+ Each daily shard includes:
25
+
26
+ - `train.txt`, `valid.txt`, `test.txt`: article-level text splits.
27
+ - `train.bpe`, `valid.bpe`, `test.bpe`: GPT-2 BPE-tokenized text.
28
+ - `bin/`: fairseq mmap binaries built from the BPE files.
29
+ - `net/train.*`, `net/test.*`: small held-out subsets used by the MoNIM auxiliary selection network.
30
+ - `net/total/`: combined auxiliary-network inputs and their fairseq binaries.
31
+ - `freq_cache_id.pickle`, `fertility_cache_id.pickle`: cached token statistics used by the reproduction code.
32
+
33
+ Top-level tokenizer files:
34
+
35
+ - `dict.txt`
36
+ - `encoder.json`
37
+ - `vocab.bpe`
38
+
39
+ Trained model checkpoints, KNN datastore/index artifacts, run logs, watcher scripts, and experiment outputs are not part of this dataset repository.
40
+
41
+ ## Data Source
42
+
43
+ The source corpus is WMT News Crawl:
44
+
45
+ - News Crawl index: https://data.statmt.org/news-crawl/
46
+ - News Crawl README: https://data.statmt.org/news-crawl/README
47
+
48
+ The News Crawl README describes the corpus as monolingual text extracted from online newspapers and released for WMT shared tasks. It also states that News Crawl does not own the extracted text and that only the packaging is released under Creative Commons CC0. Users of this repository must comply with the original News Crawl terms, the upstream take-down policy, and any applicable rights of the original text sources.
49
+
50
+ ## Preparation
51
+
52
+ The prepared daily data follows the paper setup: approximately 98% of articles are used for training, 1% for validation, and 1% for test. Text is tokenized with a GPT-2 BPE tokenizer, then preprocessed into fairseq mmap binaries with the shared dictionary.
53
+
54
+ ## Intended Use
55
+
56
+ This repository is intended for academic research and reproduction of the MoNIM continual-learning experiments. It is not intended as a general-purpose news corpus release.
57
+
58
+ ## Citation
59
+
60
+ If you use this dataset, cite the MoNIM paper and the upstream News Crawl/WMT data source:
61
+
62
+ ```bibtex
63
+ @inproceedings{peng-etal-2025-learn,
64
+ title = "Learn to Memorize: Scalable Continual Learning in Semiparametric Models with Mixture-of-Neighbors Induction Memory",
65
+ author = "Peng, Guangyue and
66
+ Ge, Tao and
67
+ Luo, Wen and
68
+ Li, Wei and
69
+ Wang, Houfeng",
70
+ booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
71
+ month = jul,
72
+ year = "2025",
73
+ address = "Vienna, Austria",
74
+ publisher = "Association for Computational Linguistics",
75
+ url = "https://aclanthology.org/2025.acl-long.1385/",
76
+ doi = "10.18653/v1/2025.acl-long.1385",
77
+ pages = "28517--28531",
78
+ }
79
+ ```
daily/1/bin/dict.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2eb4374487a7a0db6bbc4a434740d0b47ff58f180d1769a02641e0008db97fa6
3
+ size 608020
daily/1/bin/preprocess.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36c687982a686e11d8f71d844309bcb774f28366334087b2ca72f942af2bc6d0
3
+ size 1665
daily/1/bin/test.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1272a7c1639ec51defb615beac2e9f6dd287da57975e8410b091ad69b09270a7
3
+ size 55390
daily/1/bin/test.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eac151597cd76dc19e69933e44a7f10e55c9732e3e95ce3e7b83fd567fb51da1
3
+ size 14102
daily/1/bin/train.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45acb52d8b6020926e393d365979adad5eed0d56cb145f03295cf868d06376ea
3
+ size 6546440
daily/1/bin/train.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b307fedf6d0b3ded08361faadc36d3bb12ef2aeb77a613dd3db15bff18124cd
3
+ size 1610546
daily/1/bin/valid.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4c39fc3ba581f1ac6f78e19cb6604485a9f9a1e0f9b15ad23404e2697bbb561
3
+ size 56444
daily/1/bin/valid.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e3701b2c4db84406aa0e5ad970a32590d7aae73187e6b0b133389ecdfbcd137
3
+ size 14258
daily/1/fertility_cache_id.pickle ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c415051599ace90795b9d6a73adedc43276b4fb746a299af87d4c16bcd429969
3
+ size 1010036
daily/1/freq_cache_id.pickle ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09cd4fc47bf96ea28d3d5f04393dd7fddbc5b6c13f186c0084baab13ab51f779
3
+ size 1010036
daily/1/net/test.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b096783714717a46046d806039704de626fe8664dae7f59b6f3f287898568ef
3
+ size 16949
daily/1/net/test.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31403a22a5a954caa70b1d5f55944c76da6a0750b2b231097fb81a2669896d97
3
+ size 12311
daily/1/net/total/bin/dict.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2eb4374487a7a0db6bbc4a434740d0b47ff58f180d1769a02641e0008db97fa6
3
+ size 608020
daily/1/net/total/bin/preprocess.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c508e67b35889caeebd263a539214d72a861ca41fd6f8ec0fca5de1a844d429c
3
+ size 1498
daily/1/net/total/bin/test.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a48b2498e074979173a07b763129bce9343d89e75c5757eadc570b186f770d6c
3
+ size 5810
daily/1/net/total/bin/test.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71cff6d163921ba656b7af88b48f1c7121bec7d23e4374515d51c728606574c8
3
+ size 1742
daily/1/net/total/bin/train.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3006588782b972a736722f11ac1cfde375072b8b9f02b1312cf5f268fe4c110e
3
+ size 8858
daily/1/net/total/bin/train.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e02bef6ae02258b78245606df3b0bdffd0a6be94f4f1ead3b04cf05a2c85cff
3
+ size 2318
daily/1/net/total/test.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b096783714717a46046d806039704de626fe8664dae7f59b6f3f287898568ef
3
+ size 16949
daily/1/net/total/train.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f2d4d84c0f2c05aae694b5b48aa47fcd9bba7b647f233efe5fd5acb7518670a
3
+ size 26658
daily/1/net/train.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f2d4d84c0f2c05aae694b5b48aa47fcd9bba7b647f233efe5fd5acb7518670a
3
+ size 26658
daily/1/net/train.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dad5fc860deda9bca050bb6a00eac923be419278f4d9d06c1e456d4f4d24ae23
3
+ size 19449
daily/1/test.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd211de61674aca2942e4e15fb9105bdf89c251781f5542b171392b10da9913a
3
+ size 167772
daily/1/test.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a9cf64bb34c82a1e8e4a43ab3b3af2e597947bdec38f01c0d9c88529ab2e3ab
3
+ size 122815
daily/1/train.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2da5ea6924d9b132065ec6cc93684c12dd180525d124c871dee969c2d7050759
3
+ size 19966635
daily/1/train.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3565a8873515be5137a1baf50fafcb1ffb2671d5a005ae41e484bb660fc9a58c
3
+ size 14620177
daily/1/valid.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5626e547cd3588be9d131d06ce0cc17a3f1d319526a568968679c25d100a48e2
3
+ size 167805
daily/1/valid.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cc03daa1760d872c444a7e98526bc9f853e0ee0a07763dcc274be73bc399969
3
+ size 122008
daily/10/bin/dict.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2eb4374487a7a0db6bbc4a434740d0b47ff58f180d1769a02641e0008db97fa6
3
+ size 608020
daily/10/bin/preprocess.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9dc58f24ee175a73b8154a18f893e073eb33556d3caad7a41a73d3a3bea21a95
3
+ size 1674
daily/10/bin/test.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d248fbca234dec0087ffebf561fa4da225a8cefb1bef95e3906d686f62ec103
3
+ size 81590
daily/10/bin/test.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be2d79d1dcea5266207fb314585161957e245a813e9cef4cbb4aa15d7627497f
3
+ size 18866
daily/10/bin/train.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41fd8a1680bbec9de22c86a2232403f49eab474f220c8fdd2ff89f1e8d209ce7
3
+ size 7150396
daily/10/bin/train.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d9a1647739b7b64f413440343c99d175713b4e577e620b0e3a4ab08b1ea56bd
3
+ size 1695254
daily/10/bin/valid.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:441038bbc268a8186a5691890ef3bccea2d22e33f3b433f64fe8f81849993a5e
3
+ size 65520
daily/10/bin/valid.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29361e84cf48782d257fe665367b8c541cbfc237447d3bce34ebe4f51e8c78f7
3
+ size 15290
daily/10/fertility_cache_id.pickle ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:774faae88d8a956a625ad30a00f6fb8de782a4e5ecb39c1b62b65afc6b0b0781
3
+ size 1152299
daily/10/freq_cache_id.pickle ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35af115877eaee6f6d756cf0ef0b727b07adaa3dcbde3b3c04a8a2a00fdbd323
3
+ size 1152299
daily/10/net/test.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a07f3e982e59e521d3daeb3f0ac5c2abe4b1ac133b8ab55dd0954acf5443f4cc
3
+ size 30242
daily/10/net/test.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9ec4409bc43709b1f9d5f482e1e55b004f87c47a1eb97a8142c112fba8766dc
3
+ size 22206
daily/10/net/total/bin/dict.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2eb4374487a7a0db6bbc4a434740d0b47ff58f180d1769a02641e0008db97fa6
3
+ size 608020
daily/10/net/total/bin/preprocess.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6aec776e096b35f7c0e348ce8d26a70d207ac864a98a9844377d424e1568e51f
3
+ size 1509
daily/10/net/total/bin/test.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eaeac8dcc253d5929b51d75cf48a83b914c3bb0603bef2bf39f5f26fa16faf31
3
+ size 81770
daily/10/net/total/bin/test.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb3deaa7a1702d6a415e4424f31fef3b9abb252d8b8167ab36e39bf24f470be3
3
+ size 18938
daily/10/net/total/bin/train.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b72024e4c62d3af982ceebd002e988ade9b18b9e0f0a04805bf75d4fc1c39464
3
+ size 141964
daily/10/net/total/bin/train.idx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c704cb60f6e3c2064e6b9b37f1bf32a58b2c3cbb90422c2873bbd47c5c59c3a
3
+ size 34298
daily/10/net/total/test.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92277b40f5aee8a301d12dcf871e4af0701da7a01c3f8fc396b13a2f0c2f9834
3
+ size 250981
daily/10/net/total/train.bpe ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:766d8c83f32040eacca4fc2b4c7192e334efac3a83dae80194afd07b1e73b4fa
3
+ size 436086