Hugging Face's logo

Datasets:
ejschwartz
/
idioms-realtype

Modalities:
Text
Formats:
json
Size:
100K - 1M
ArXiv:
Tags:
datasets
Libraries:
Datasets
pandas
Polars
Dataset card Data Studio Files
xet
 Community
2
ejschwartz commited on
Commit
031316d
·
1 Parent(s): f558a20

Add hex-rays-parity configs to the dataset

Browse files
Files changed (33) hide show
  1. README.md +80 -10
  2. by-binary/test.jsonl → by-binary-hex-rays-parity-O0/test.json.zst +2 -2
  3. by-binary-hex-rays-parity-O0/train.json.zst +3 -0
  4. by-function/valid.json → by-binary-hex-rays-parity-O0/valid.json.zst +2 -2
  5. by-binary/valid.jsonl → by-binary-hex-rays-parity-O1_noinline/test.json.zst +2 -2
  6. by-binary-hex-rays-parity-O1_noinline/train.json.zst +3 -0
  7. by-function/test.json → by-binary-hex-rays-parity-O1_noinline/valid.json.zst +2 -2
  8. by-binary-hex-rays-parity-O2_noinline/test.json.zst +3 -0
  9. by-binary-hex-rays-parity-O2_noinline/train.json.zst +3 -0
  10. by-binary-hex-rays-parity-O2_noinline/valid.json.zst +3 -0
  11. by-binary-hex-rays-parity-O3_noinline/test.json.zst +3 -0
  12. by-binary-hex-rays-parity-O3_noinline/train.json.zst +3 -0
  13. by-binary-hex-rays-parity-O3_noinline/valid.json.zst +3 -0
  14. by-binary/test.json.zst +3 -0
  15. by-binary/train.json.zst +3 -0
  16. by-binary/train.jsonl +0 -3
  17. by-binary/valid.json.zst +3 -0
  18. by-function-hex-rays-parity-O0/test.json.zst +3 -0
  19. by-function-hex-rays-parity-O0/train.json.zst +3 -0
  20. by-function-hex-rays-parity-O0/valid.json.zst +3 -0
  21. by-function-hex-rays-parity-O1_noinline/test.json.zst +3 -0
  22. by-function-hex-rays-parity-O1_noinline/train.json.zst +3 -0
  23. by-function-hex-rays-parity-O1_noinline/valid.json.zst +3 -0
  24. by-function-hex-rays-parity-O2_noinline/test.json.zst +3 -0
  25. by-function-hex-rays-parity-O2_noinline/train.json.zst +3 -0
  26. by-function-hex-rays-parity-O2_noinline/valid.json.zst +3 -0
  27. by-function-hex-rays-parity-O3_noinline/test.json.zst +3 -0
  28. by-function-hex-rays-parity-O3_noinline/train.json.zst +3 -0
  29. by-function-hex-rays-parity-O3_noinline/valid.json.zst +3 -0
  30. by-function/test.json.zst +3 -0
  31. by-function/train.json +0 -3
  32. by-function/train.json.zst +3 -0
  33. by-function/valid.json.zst +3 -0
README.md CHANGED
@@ -1,21 +1,85 @@
1
  ---
2
  configs:
3
- - config_name: by_function
4
  data_files:
5
  - split: train
6
- path: "by-function/train.json"
7
  - split: test
8
- path: "by-function/test.json"
9
  - split: validate
10
- path: "by-function/valid.json"
11
- - config_name: by_binary
12
  data_files:
13
  - split: train
14
- path: "by-binary/train.jsonl"
15
  - split: test
16
- path: "by-binary/test.jsonl"
17
  - split: validate
18
- path: "by-binary/valid.jsonl"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
19
  ---
20
 
21
  # Dataset Card for REALTYPE
@@ -44,7 +108,7 @@ The REALTYPE dataset is primarily intended for training and evaluating neural de
44
 
45
  ## Dataset Structure
46
 
47
- The REALTYPE dataset contains 154,301 training functions, 540 validation functions, and 2,322 test functions, all extracted from C code repositories on GitHub. Each example in the dataset consists of:
48
 
49
  * Decompiled code: The output of running the Hex-Rays decompiler on compiled binaries.
50
  * Original source code: The canonicalized form of the original function from source.
@@ -53,6 +117,12 @@ The REALTYPE dataset contains 154,301 training functions, 540 validation functio
53
 
54
  The original REALTYPE dataset was only organized by binary. This view can be accessed using the `by_binary` configuration. This dataset also contains a `by_function` configuration, which organizes the data by function. Most data is in both views, but the `by_function` view does not contain the call graph or unmatched functions (decompiled functions for which the original source code could not be found).
55
 
 
 
 
 
 
 
56
  ## Dataset Creation
57
 
58
  ### Curation Rationale
@@ -89,4 +159,4 @@ The REALTYPE dataset has several technical limitations acknowledged in the paper
89
 
90
  * Luke Dramko: Carnegie Mellon University (lukedram@cs.cmu.edu)
91
  * Claire Le Goues: Carnegie Mellon University (clegoues@cs.cmu.edu)
92
- * Edward J. Schwartz: Carnegie Mellon University Software Engineering Institute (eschwartz@cert.org)
 
1
  ---
2
  configs:
3
+ - config_name: by-function
4
  data_files:
5
  - split: train
6
+ path: "by-function/train.json.zst"
7
  - split: test
8
+ path: "by-function/test.json.zst"
9
  - split: validate
10
+ path: "by-function/valid.json.zst"
11
+ - config_name: by-binary
12
  data_files:
13
  - split: train
14
+ path: "by-binary/train.json.zst"
15
  - split: test
16
+ path: "by-binary/test.json.zst"
17
  - split: validate
18
+ path: "by-binary/valid.json.zst"
19
+ - config_name: by-binary-hex-rays-parity-O0
20
+ data_files:
21
+ - split: train
22
+ path: "by-binary-hex-rays-parity-O0/train.json.zst"
23
+ - split: test
24
+ path: "by-binary-hex-rays-parity-O0/test.json.zst"
25
+ - split: validate
26
+ path: "by-binary-hex-rays-parity-O0/valid.json.zst"
27
+ - config_name: by-binary-hex-rays-parity-O1_noinline
28
+ data_files:
29
+ - split: train
30
+ path: "by-binary-hex-rays-parity-O1_noinline/train.json.zst"
31
+ - split: test
32
+ path: "by-binary-hex-rays-parity-O1_noinline/test.json.zst"
33
+ - split: validate
34
+ path: "by-binary-hex-rays-parity-O1_noinline/valid.json.zst"
35
+ - config_name: by-binary-hex-rays-parity-O2_noinline
36
+ data_files:
37
+ - split: train
38
+ path: "by-binary-hex-rays-parity-O2_noinline/train.json.zst"
39
+ - split: test
40
+ path: "by-binary-hex-rays-parity-O2_noinline/test.json.zst"
41
+ - split: validate
42
+ path: "by-binary-hex-rays-parity-O2_noinline/valid.json.zst"
43
+ - config_name: by-binary-hex-rays-parity-O3_noinline
44
+ data_files:
45
+ - split: train
46
+ path: "by-binary-hex-rays-parity-O3_noinline/train.json.zst"
47
+ - split: test
48
+ path: "by-binary-hex-rays-parity-O3_noinline/test.json.zst"
49
+ - split: validate
50
+ path: "by-binary-hex-rays-parity-O3_noinline/valid.json.zst"
51
+ - config_name: by-function-hex-rays-parity-O0
52
+ data_files:
53
+ - split: train
54
+ path: "by-function-hex-rays-parity-O0/train.json.zst"
55
+ - split: test
56
+ path: "by-function-hex-rays-parity-O0/test.json.zst"
57
+ - split: validate
58
+ path: "by-function-hex-rays-parity-O0/valid.json.zst"
59
+ - config_name: by-function-hex-rays-parity-O1_noinline
60
+ data_files:
61
+ - split: train
62
+ path: "by-function-hex-rays-parity-O1_noinline/train.json.zst"
63
+ - split: test
64
+ path: "by-function-hex-rays-parity-O1_noinline/test.json.zst"
65
+ - split: validate
66
+ path: "by-function-hex-rays-parity-O1_noinline/valid.json.zst"
67
+ - config_name: by-function-hex-rays-parity-O2_noinline
68
+ data_files:
69
+ - split: train
70
+ path: "by-function-hex-rays-parity-O2_noinline/train.json.zst"
71
+ - split: test
72
+ path: "by-function-hex-rays-parity-O2_noinline/test.json.zst"
73
+ - split: validate
74
+ path: "by-function-hex-rays-parity-O2_noinline/valid.json.zst"
75
+ - config_name: by-function-hex-rays-parity-O3_noinline
76
+ data_files:
77
+ - split: train
78
+ path: "by-function-hex-rays-parity-O3_noinline/train.json.zst"
79
+ - split: test
80
+ path: "by-function-hex-rays-parity-O3_noinline/test.json.zst"
81
+ - split: validate
82
+ path: "by-function-hex-rays-parity-O3_noinline/valid.json.zst"
83
  ---
84
 
85
  # Dataset Card for REALTYPE
 
108
 
109
  ## Dataset Structure
110
 
111
+ The REALTYPE dataset contains 154,301 training functions, 540 validation functions, and 2,322 test functions, all extracted from C code repositories on GitHub. The base (by-binary) splits contain data for 33,500 training binaries, 109 validation binaries, and 400 test binaries. Each example in the dataset consists of:
112
 
113
  * Decompiled code: The output of running the Hex-Rays decompiler on compiled binaries.
114
  * Original source code: The canonicalized form of the original function from source.
 
117
 
118
  The original REALTYPE dataset was only organized by binary. This view can be accessed using the `by_binary` configuration. This dataset also contains a `by_function` configuration, which organizes the data by function. Most data is in both views, but the `by_function` view does not contain the call graph or unmatched functions (decompiled functions for which the original source code could not be found).
119
 
120
+ ### Parity subset
121
+
122
+ The dataset also includes parity subsets for binary/function parity experiments available in configurations such as `by-binary-hex-rays-parity-O0`, `by-binary-hex-rays-parity-O1_noinline`, `by-binary-hex-rays-parity-O2_noinline`, `by-binary-hex-rays-parity-O3_noinline`, and their `by-function-hex-rays-parity-*` counterparts; these contain matched binaries and functions across optimization levels and inlining variants. The paper describes parity datasets for both Hex-Rays and Ghidra, but this repository's parity subsets are for Hex-Rays only. As described in section 3.3 of the Idioms paper, the parity subset includes only functions that were successfully decompiled by Hex-Rays at all included optimization levels/variants (i.e., functions present in every binary variant). See the Idioms paper (https://arxiv.org/abs/2502.04536, section 3.3) and the Zenodo record (https://zenodo.org/records/15683630) for details and citation.
123
+
124
+ Across the parity subsets organized by function (the `by-function-hex-rays-parity-*` configurations) there are 106,238 training functions (≈68.9% of the full training split), 431 validation functions (≈79.8%), and 1,818 test functions (≈78.3%) present. The corresponding parity subsets organized by binary (the `by-binary-hex-rays-parity-*` configurations) contain 23,897 training binaries (≈71.4%), 88 validation binaries (≈80.7%), and 337 test binaries (≈84.3%).
125
+
126
  ## Dataset Creation
127
 
128
  ### Curation Rationale
 
159
 
160
  * Luke Dramko: Carnegie Mellon University (lukedram@cs.cmu.edu)
161
  * Claire Le Goues: Carnegie Mellon University (clegoues@cs.cmu.edu)
162
+ * Edward J. Schwartz: Carnegie Mellon University Software Engineering Institute (eschwartz@cert.org)
by-binary/test.jsonl → by-binary-hex-rays-parity-O0/test.json.zst RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6e81ce24698f35f09f239da804e0816bb4a3d28e389519361e86a9890ba96643
3
- size 8584988
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e6b2ddafd71f92bd62dba32884effc751b73a71c3b096d5f97bcaaf384a0a80
3
+ size 638537
by-binary-hex-rays-parity-O0/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cd4d719c5b33a1421cd18986c0bcfc8f1e4e8d13bf594fcce3aa496b345d942
3
+ size 41272400
by-function/valid.json → by-binary-hex-rays-parity-O0/valid.json.zst RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c2722357a369d33b632bc1af0de6063aeec8d50a0d9f2be2390189399f7a241c
3
- size 1678474
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e4c89737baed2b9b297bd524712288d4bf6c0e869331b2633a94998ae798b8f
3
+ size 194848
by-binary/valid.jsonl → by-binary-hex-rays-parity-O1_noinline/test.json.zst RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d043a68629a19276c88ea949b23647113e63adbd54a1bfcfb2706640b271cfbc
3
- size 1815308
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b87ef8e15a04bd41414849a7d047356e3a72eecc76fe0131e6de10dd2dc4249c
3
+ size 658493
by-binary-hex-rays-parity-O1_noinline/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84eef67f1d0a238527ea08773b8cfdc10949c00acd4ef6957d8fe28d98c98ff8
3
+ size 42555069
by-function/test.json → by-binary-hex-rays-parity-O1_noinline/valid.json.zst RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:38e5845002c98796866b9445686730a7e13d22f8f7700b5de409a3d33293335c
3
- size 8388877
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79c6f441cc3c74368ef7876fedaab78f565bbc5d3b9167ce3e49256a4473d056
3
+ size 203047
by-binary-hex-rays-parity-O2_noinline/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:360da2270974093ed8d7018f677d2eb6668cacc3515c080b7ad8e377146fe3f6
3
+ size 676292
by-binary-hex-rays-parity-O2_noinline/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa0e392bce18aa34ee86d9b85a14b13abbdd12711596f9df17f7dcbaafd302a5
3
+ size 43923453
by-binary-hex-rays-parity-O2_noinline/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d30a02cb8b42e840bb87f3dfabba0761b41773a93808dd12b2008dc550f5b42e
3
+ size 207166
by-binary-hex-rays-parity-O3_noinline/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa88560a5fe7dc9cb947b475e9caac088b9d872eeb421e6da4eb086c044dad49
3
+ size 682028
by-binary-hex-rays-parity-O3_noinline/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2b84c9e973937b0bc9b24f1fd9322c85392568dc45d743e4851a4bf164285e3
3
+ size 44989871
by-binary-hex-rays-parity-O3_noinline/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d35ac237421156f391a670c11781fb3ba7282fdddfa74fca187c8c49d54e3b2f
3
+ size 210676
by-binary/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d932b1aa89fc5d15d8fc988cb6db93bae2a12e7e70ad3aaeecae90eaaeb79b7
3
+ size 802051
by-binary/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d1ebf66c781192b9bb65e615c44d9b3b54756997d74caddcb553d63e4eea54e
3
+ size 63372430
by-binary/train.jsonl DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:93289333501536d30cbe46fb66029ca3dc005f2aa24f64de08c0eaf550c2e6e3
3
- size 644285652
 
 
 
 
by-binary/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b71d49e1de130b064b141692b21d466c1b69315c84ddb424f5c902949fbd067
3
+ size 244788
by-function-hex-rays-parity-O0/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce6b53fc2a1ed6ea22d53188e4b844db1bff37bd8b50d1234700e205f62a6099
3
+ size 530984
by-function-hex-rays-parity-O0/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a5a425f5cc6ba84c88e0e63d8bf2a529b57e8156323f5a0cba828492f84cd05
3
+ size 33712812
by-function-hex-rays-parity-O0/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cffe6621a008c7725ac61bff3bc6beb92315b30fab427c0b5b8123ef1046563a
3
+ size 163963
by-function-hex-rays-parity-O1_noinline/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:453ffe3360c4ed4216bf26230ed6c661ac1b7ca35175fc940ee98821aeeb21e4
3
+ size 545111
by-function-hex-rays-parity-O1_noinline/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bf33090f32d7e08be1401300c2ff4ac1814311edadfa34f0644d8d77b3450c9
3
+ size 34678111
by-function-hex-rays-parity-O1_noinline/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d7f1cefa28a6b8a141e1f34dcc47c8d1adbb935d83cc0d99e1e16f920ba2840
3
+ size 169438
by-function-hex-rays-parity-O2_noinline/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73cb04cc84ff3dcd5da7c96c6f3c6dfc566bab2ff07e4b8825a15a43e034bb7c
3
+ size 560671
by-function-hex-rays-parity-O2_noinline/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46e544e870d60e09faaeb7ca87a9bb312ffb6d6bb7d91cb1f7b54a0d99b78a3c
3
+ size 35689932
by-function-hex-rays-parity-O2_noinline/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8fc7597b971ce272a01a4e3bb3f11001eeb09faacb22d71345507f0c18c4eb1
3
+ size 173598
by-function-hex-rays-parity-O3_noinline/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8098f7c996e869029dfa2fb07676252c7af59a4391a698fdc3b426ce803e94d4
3
+ size 569710
by-function-hex-rays-parity-O3_noinline/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd47ec396c308a4c0ec6edc99d50ca80b959bc8ed82eb75601fd7a7d988216c5
3
+ size 36467211
by-function-hex-rays-parity-O3_noinline/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4666fab741a04fdfae97ed3060d2c87f89a09f5d10f8b9025709dce5f875d432
3
+ size 175672
by-function/test.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6dee1a9daa86dfba42dbd7900c5fa14b2882896d817b435e373b82e86296c7f2
3
+ size 717136
by-function/train.json DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:ba8b4b1c59903ffa4fa81be240a8d1a559b25fe734f26d829a3b98cdb099626b
3
- size 621928174
 
 
 
 
by-function/train.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b8a1abef4a82308ec4704e530d929b196f3d5d799d2ded06bd628c2742ef09a
3
+ size 55496260
by-function/valid.json.zst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:998e0aab2c377f39ae45cf3278a7e5ed8a0327137882f7040c9252716aaf9f42
3
+ size 209045