ddiruscio/logseq
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
ded48682a1b286eec63f153e33a7c9f2e38b26bf
logseq/assets/TOSEM-2023-0288_Proof_hi_1693841825871_0.edn
T
ddiruscio 97fba407ce Major2
2025-06-05 22:07:12 +02:00

2595 lines
142 KiB
Clojure
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
{:highlights [{:id #uuid "65198f24-5b64-43dc-abe5-a35bf443753e",
:page 2,
:position {:bounding {:x1 183.0032730102539,
:y1 384.5658874511719,
:x2 843.9134292602539,
:y2 400.0846862792969,
:width 1060.8,
:height 1372.8},
:rects ({:x1 183.0032730102539,
:y1 384.5658874511719,
:x2 843.9134292602539,
:y2 400.0846862792969,
:width 1060.8,
:height 1372.8}),
:page 2},
:content {:text "deep learning-based techniques have shown promising performance on various tasks related to software engineering"},
:properties {:color "green"}}
{:id #uuid "65198f61-6aab-4f71-84b8-04a3c5a51746",
:page 2,
:position {:bounding {:x1 173.40627670288086,
:y1 492.47007751464844,
:x2 321.5644950866699,
:y2 507.98887634277344,
:width 1060.8,
:height 1372.8},
:rects ({:x1 173.40627670288086,
:y1 492.47007751464844,
:x2 321.5644950866699,
:y2 507.98887634277344,
:width 1060.8,
:height 1372.8}),
:page 2},
:content {:text "comment updating dataset"},
:properties {:color "yellow"}}
{:id #uuid "6519908b-670b-4c0f-bfa7-6cf084367f32",
:page 2,
:position {:bounding {:x1 0,
:y1 773.0589904785156,
:x2 736.0829162597656,
:y2 1313.8731689453125,
:width 897.6,
:height 1161.6000000000001},
:rects ({:x1 516.1744384765625,
:y1 773.0589904785156,
:x2 736.0829162597656,
:y2 787.664924621582,
:width 897.6,
:height 1161.6000000000001}
{:x1 107.70925426483154,
:y1 793.1564178466797,
:x2 256.256236076355,
:y2 807.7623443603516,
:width 897.6,
:height 1161.6000000000001}
{:x1 0,
:y1 1293.7900390625,
:x2 0,
:y2 1313.8731689453125,
:width 897.6,
:height 1161.6000000000001}),
:page 2},
:content {:text "High-quality data is a key factor in deep learning-based approaches"},
:properties {:color "green"}}
{:id #uuid "65199206-3022-4f58-9825-618498bc74da",
:page 3,
:position {:bounding {:x1 176.19834518432617,
:y1 541.6700134277344,
:x2 282.8702201843262,
:y2 556.2759399414062,
:width 897.6,
:height 1161.6000000000001},
:rects ({:x1 176.19834518432617,
:y1 541.6700134277344,
:x2 282.8702201843262,
:y2 556.2759399414062,
:width 897.6,
:height 1161.6000000000001}),
:page 3},
:content {:text "Comment updating "},
:properties {:color "blue"}}
{:id #uuid "65199218-a454-4b39-997a-031ea97f4c0b",
:page 3,
:position {:bounding {:x1 161.20729064941406,
:y1 561.7626647949219,
:x2 782.5102081298828,
:y2 576.3685913085938,
:width 897.6,
:height 1161.6000000000001},
:rects ({:x1 161.20729064941406,
:y1 561.7626647949219,
:x2 782.5102081298828,
:y2 576.3685913085938,
:width 897.6,
:height 1161.6000000000001}),
:page 3},
:content {:text "This task aims to automatically update the corresponding comments based on code changes made by developer"},
:properties {:color "purple"}}
{:id #uuid "65199327-ba6b-4525-939a-dcc07b4652b5",
:page 3,
:position {:bounding {:x1 0,
:y1 929.9250030517578,
:x2 1223.4700164794922,
:y2 983.78125,
:width 1387.2,
:height 1795.2},
:rects ({:x1 711.0892791748047,
:y1 929.9250030517578,
:x2 1223.4700164794922,
:y2 952.7249984741211,
:width 1387.2,
:height 1795.2}
{:x1 0,
:y1 958,
:x2 0,
:y2 977.6000061035156,
:width 1387.2,
:height 1795.2}
{:x1 249.6937484741211,
:y1 960.9812469482422,
:x2 598.4816513061523,
:y2 983.78125,
:width 1387.2,
:height 1795.2}),
:page 3},
:content {:text " comment updating can take into account more information, such as code changes and old comment"},
:properties {:color "purple"}}
{:id #uuid "65199337-7123-4dc5-a6c8-23f5536673e2",
:page 3,
:position {:bounding {:x1 0,
:y1 989.9999847412109,
:x2 1221.2338409423828,
:y2 1045.8999938964844,
:width 1387.2,
:height 1795.2},
:rects ({:x1 0,
:y1 989.9999847412109,
:x2 0,
:y2 1009.5999908447266,
:width 1387.2,
:height 1795.2}
{:x1 646.7042999267578,
:y1 992.0374908447266,
:x2 1221.2338409423828,
:y2 1014.8374938964844,
:width 1387.2,
:height 1795.2}
{:x1 249.6937484741211,
:y1 1023.0999908447266,
:x2 498.48082733154297,
:y2 1045.8999938964844,
:width 1387.2,
:height 1795.2}),
:page 3},
:content {:text " if comments are not updated in a timely manner, they may mislead future development activities."},
:properties {:color "purple"}}
{:id #uuid "6519934c-807c-464b-a598-c43531e6dcea",
:page 3,
:position {:bounding {:x1 0,
:y1 1005.9999847412109,
:x2 1221.6533660888672,
:y2 1108.0124969482422,
:width 1387.2,
:height 1795.2},
:rects ({:x1 0,
:y1 1005.9999847412109,
:x2 0,
:y2 1025.5999908447266,
:width 1387.2,
:height 1795.2}
{:x1 0,
:y1 1021.9999847412109,
:x2 0,
:y2 1041.5999908447266,
:width 1387.2,
:height 1795.2}
{:x1 615.9065399169922,
:y1 1023.0999908447266,
:x2 1221.6533660888672,
:y2 1045.8999938964844,
:width 1387.2,
:height 1795.2}
{:x1 249.6937484741211,
:y1 1054.156234741211,
:x2 1221.0645904541016,
:y2 1076.9562377929688,
:width 1387.2,
:height 1795.2}
{:x1 249.6937484741211,
:y1 1085.2124938964844,
:x2 876.8065414428711,
:y2 1108.0124969482422,
:width 1387.2,
:height 1795.2}),
:page 3},
:content {:text " if obsolete TODO comments are not removed, they may introduce new bugs in future development [ 6]. For the task of comment updating, the data we expect is to be able to meet the real editing intention of developers and reflect the changes in code functionality"},
:properties {:color "green"}}
{:id #uuid "651993fb-82cc-4432-a5eb-2bdc37dee2c0",
:page 3,
:position {:bounding {:x1 0,
:y1 1069.999984741211,
:x2 1221.2401580810547,
:y2 1201.1812438964844,
:width 1387.2,
:height 1795.2},
:rects ({:x1 0,
:y1 1069.999984741211,
:x2 0,
:y2 1089.5999908447266,
:width 1387.2,
:height 1795.2}
{:x1 730.8583831787109,
:y1 1147.3249816894531,
:x2 1221.2401580810547,
:y2 1170.125,
:width 1387.2,
:height 1795.2}
{:x1 249.6937484741211,
:y1 1178.3812561035156,
:x2 845.7435531616211,
:y2 1201.1812438964844,
:width 1387.2,
:height 1795.2}),
:page 3},
:content {:text "Besides, comment updating datasets are often crawled from online repositories, such as Github, by extracting commit history versions"},
:properties {:color "green"}}
{:id #uuid "65199415-7d85-4913-bddf-e36fe63a8b31",
:page 3,
:position {:bounding {:x1 0,
:y1 1101.999984741211,
:x2 1221.0216522216797,
:y2 1263.2937316894531,
:width 1387.2,
:height 1795.2},
:rects ({:x1 0,
:y1 1101.999984741211,
:x2 0,
:y2 1121.6000061035156,
:width 1387.2,
:height 1795.2}
{:x1 1040.2836151123047,
:y1 1209.4375,
:x2 1221.0216522216797,
:y2 1232.2374877929688,
:width 1387.2,
:height 1795.2}
{:x1 249.6937484741211,
:y1 1240.4937438964844,
:x2 480.5791244506836,
:y2 1263.2937316894531,
:width 1387.2,
:height 1795.2}),
:page 3},
:content {:text "a noisy data sample in the comment updating task."},
:properties {:color "green"}}
{:id #uuid "6522dbea-ac76-4447-bef2-71012907b112",
:page 3,
:position {:bounding {:x1 0,
:y1 980.4375915527344,
:x2 897.4771041870117,
:y2 1185.5553894042969,
:width 1019.9999999999999,
:height 1319.9999999999998},
:rects ({:x1 769.1977462768555,
:y1 980.4375915527344,
:x2 897.4771041870117,
:y2 997.3257141113281,
:width 1019.9999999999999,
:height 1319.9999999999998}
{:x1 183.5940170288086,
:y1 1003.2736511230469,
:x2 665.9337692260742,
:y2 1020.1617736816406,
:width 1019.9999999999999,
:height 1319.9999999999998}
{:x1 0,
:y1 1165.4722595214844,
:x2 0,
:y2 1185.5553894042969,
:width 1019.9999999999999,
:height 1319.9999999999998}),
:page 3},
:content {:text "From a code change perspective, the comment modification does not reflect the changes in the code."},
:properties {:color "green"}}
{:id #uuid "6522dc22-f58c-401d-b6a7-3076fff17688",
:page 3,
:position {:bounding {:x1 0,
:y1 1048.9457397460938,
:x2 897.9970321655273,
:y2 1233.5453491210938,
:width 1019.9999999999999,
:height 1319.9999999999998},
:rects ({:x1 266.21578216552734,
:y1 1048.9457397460938,
:x2 897.9970321655273,
:y2 1065.8338317871094,
:width 1019.9999999999999,
:height 1319.9999999999998}
{:x1 183.5940170288086,
:y1 1071.7817687988281,
:x2 407.1935043334961,
:y2 1088.6698608398438,
:width 1019.9999999999999,
:height 1319.9999999999998}
{:x1 0,
:y1 1213.4621887207031,
:x2 0,
:y2 1233.5453491210938,
:width 1019.9999999999999,
:height 1319.9999999999998}),
:page 3},
:content {:text " If the dataset contains a large number of such noisy data, the results generated by a model trained with the dataset would also be misleading"},
:properties {:color "blue"}}
{:id #uuid "6522e59e-31d8-4013-9b60-e47ff0ac8d58",
:page 4,
:position {:bounding {:x1 135.75389099121094,
:y1 438.86981201171875,
:x2 444.9194793701172,
:y2 455.301513671875,
:width 996,
:height 1288.9411764705883},
:rects ({:x1 135.75389099121094,
:y1 438.86981201171875,
:x2 444.9194793701172,
:y2 455.301513671875,
:width 996,
:height 1288.9411764705883}),
:page 4},
:content {:text "Figure 2 shows an overview of our research procedur"},
:properties {:color "green"}}
{:id #uuid "6522eb18-02cb-4003-a0d7-8581843d20f3",
:page 4,
:position {:bounding {:x1 269.71669006347656,
:y1 528.6306991577148,
:x2 604.6752471923828,
:y2 547.3445358276367,
:width 1142.3999999999999,
:height 1478.3999999999999},
:rects ({:x1 269.71669006347656,
:y1 528.6306991577148,
:x2 604.6752471923828,
:y2 547.3445358276367,
:width 1142.3999999999999,
:height 1478.3999999999999}),
:page 4},
:content {:text "weak correlation between old and new comments,"},
:properties {:color "blue"}}
{:id #uuid "6522eb21-d417-453a-9308-c169d0e02032",
:page 4,
:position {:bounding {:x1 0,
:y1 528.6306991577148,
:x2 937.1569976806641,
:y2 913.6125717163086,
:width 1142.3999999999999,
:height 1478.3999999999999},
:rects ({:x1 638.2573394775391,
:y1 528.6306991577148,
:x2 937.1569976806641,
:y2 547.3445358276367,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 137.0803985595703,
:y1 554.2053451538086,
:x2 261.53443908691406,
:y2 572.9191818237305,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 0,
:y1 893.5294418334961,
:x2 0,
:y2 913.6125717163086,
:width 1142.3999999999999,
:height 1478.3999999999999}),
:page 4},
:content {:text "weak correlation between code changes and comment changes"},
:properties {:color "blue"}}
{:id #uuid "6522eb2e-2901-4762-997a-a85be116406f",
:page 4,
:position {:bounding {:x1 195.7086181640625,
:y1 579.7871017456055,
:x2 276.1130828857422,
:y2 598.5009994506836,
:width 1142.3999999999999,
:height 1478.3999999999999},
:rects ({:x1 195.7086181640625,
:y1 579.7871017456055,
:x2 276.1130828857422,
:y2 598.5009994506836,
:width 1142.3999999999999,
:height 1478.3999999999999}),
:page 4},
:content {:text "CupCleaner"},
:properties {:color "purple"}}
{:id #uuid "6522eb52-eb79-4a98-b1ce-1320f4a48540",
:page 4,
:position {:bounding {:x1 0,
:y1 630.9364242553711,
:x2 936.8831939697266,
:y2 977.5991439819336,
:width 1142.3999999999999,
:height 1478.3999999999999},
:rects ({:x1 614.8409576416016,
:y1 630.9364242553711,
:x2 936.8831939697266,
:y2 649.6502914428711,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 137.0803985595703,
:y1 656.5110702514648,
:x2 484.2821807861328,
:y2 675.2249374389648,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 0,
:y1 957.5159530639648,
:x2 0,
:y2 977.5991439819336,
:width 1142.3999999999999,
:height 1478.3999999999999}),
:page 4},
:content {:text " The first step is to design a criterion to calculate the score of all data and map them to a distribution"},
:properties {:color "green"}}
{:id #uuid "6522eba7-0d34-41cf-a392-8053d2061aac",
:page 4,
:position {:bounding {:x1 0,
:y1 784.3914413452148,
:x2 937.1673736572266,
:y2 1073.5789413452148,
:width 1142.3999999999999,
:height 1478.3999999999999},
:rects ({:x1 335.33985900878906,
:y1 784.3914413452148,
:x2 937.1673736572266,
:y2 803.1053085327148,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 137.0803985595703,
:y1 809.9732284545898,
:x2 660.5012969970703,
:y2 828.6870956420898,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 0,
:y1 1053.4958114624023,
:x2 0,
:y2 1073.5789413452148,
:width 1142.3999999999999,
:height 1478.3999999999999}),
:page 4},
:content {:text " we design an anchor point search technique to adapt to different datasets. We are able to adaptively trim the tail of the score distribution, i.e., filter out the noisy data."},
:properties {:color "blue"}}
{:id #uuid "6522ec49-663d-43d6-ab74-c0528507ae0c",
:page 4,
:position {:bounding {:x1 0,
:y1 835.5478820800781,
:x2 937.1754302978516,
:y2 1105.5722351074219,
:width 1142.3999999999999,
:height 1478.3999999999999},
:rects ({:x1 630.3087921142578,
:y1 835.5478820800781,
:x2 937.1754302978516,
:y2 854.26171875,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 137.0803985595703,
:y1 861.1225280761719,
:x2 311.73341369628906,
:y2 879.8363952636719,
:width 1142.3999999999999,
:height 1478.3999999999999}
{:x1 0,
:y1 1085.4890441894531,
:x2 0,
:y2 1105.5722351074219,
:width 1142.3999999999999,
:height 1478.3999999999999}),
:page 4},
:content {:text "mixing the noise data and high-quality data identified by CupCleaner"},
:properties {:color "green"}}
{:id #uuid "65230fcc-e64a-4439-90ba-6980093950db",
:page 4,
:position {:bounding {:x1 0,
:y1 651.6899719238281,
:x2 669.386157989502,
:y2 1169.882080078125,
:width 816,
:height 1056},
:rects ({:x1 544.1741600036621,
:y1 651.6899719238281,
:x2 668.9670677185059,
:y2 665.0787353515625,
:width 816,
:height 1056}
{:x1 97.9196662902832,
:y1 669.9616394042969,
:x2 669.386157989502,
:y2 683.3504028320312,
:width 816,
:height 1056}
{:x1 97.9196662902832,
:y1 688.2285766601562,
:x2 233.63972854614258,
:y2 701.6173400878906,
:width 816,
:height 1056}
{:x1 0,
:y1 1133.7998657226562,
:x2 0,
:y2 1153.883056640625,
:width 816,
:height 1056}
{:x1 0,
:y1 1149.7988891601562,
:x2 0,
:y2 1169.882080078125,
:width 816,
:height 1056}),
:page 4},
:content {:text "The experimental results show that cleaning the training and validation sets with our approach can effectively improve the performance of the models on the same test set"},
:properties {:color "purple"}}
{:id #uuid "65230fe4-59f5-4925-8f7e-5c2b5283cd29",
:page 4,
:position {:bounding {:x1 0,
:y1 706.4954833984375,
:x2 669.3802757263184,
:y2 1201.8800659179688,
:width 816,
:height 1056},
:rects ({:x1 402.5672264099121,
:y1 706.4954833984375,
:x2 669.3802757263184,
:y2 719.88427734375,
:width 816,
:height 1056}
{:x1 97.9196662902832,
:y1 724.7671813964844,
:x2 340.1070137023926,
:y2 738.1559448242188,
:width 816,
:height 1056}
{:x1 0,
:y1 1181.7969360351562,
:x2 0,
:y2 1201.8800659179688,
:width 816,
:height 1056}),
:page 4},
:content {:text " the prefromance of the model trained on the cleaned dataset still improves on all performance metrics. "},
:properties {:color "green"}}
{:id #uuid "65231058-389d-460a-8848-2d7ae2c7bc02",
:page 4,
:position {:bounding {:x1 0,
:y1 761.1964416503906,
:x2 669.3837776184082,
:y2 1265.8761291503906,
:width 816,
:height 1056},
:rects ({:x1 527.9413719177246,
:y1 761.1964416503906,
:x2 669.3814964294434,
:y2 774.5851745605469,
:width 816,
:height 1056}
{:x1 97.9196662902832,
:y1 779.4681091308594,
:x2 669.3837776184082,
:y2 792.8569030761719,
:width 816,
:height 1056}
{:x1 97.9196662902832,
:y1 797.7350158691406,
:x2 237.00874710083008,
:y2 811.1238098144531,
:width 816,
:height 1056}
{:x1 0,
:y1 1229.7939758300781,
:x2 0,
:y2 1249.8771057128906,
:width 816,
:height 1056}
{:x1 0,
:y1 1245.7929992675781,
:x2 0,
:y2 1265.8761291503906,
:width 816,
:height 1056}),
:page 4},
:content {:text ". We find that the evaluation performance on the cleaned test set is significantly better than that on the pure noise test set, regardless of whether the training set is cleaned or not."},
:properties {:color "yellow"}}
{:id #uuid "6527c0ec-e2b1-4e07-a604-08cd13e590d4",
:page 4,
:position {:bounding {:x1 312.4252700805664,
:y1 1059.7432250976562,
:x2 566.0050277709961,
:y2 1079.826416015625,
:width 1224,
:height 1584},
:rects ({:x1 312.4252700805664,
:y1 1059.7432250976562,
:x2 566.0050277709961,
:y2 1079.826416015625,
:width 1224,
:height 1584}),
:page 4},
:content {:text "30% of the data in the training set"},
:properties {:color "purple"}}
{:id #uuid "6527c209-af02-417d-ace0-63b94c12fb12",
:page 5,
:position {:bounding {:x1 442.7741241455078,
:y1 303.54383850097656,
:x2 756.1725921630859,
:y2 323.6269989013672,
:width 1224,
:height 1584},
:rects ({:x1 442.7741241455078,
:y1 303.54383850097656,
:x2 756.1725921630859,
:y2 323.6269989013672,
:width 1224,
:height 1584}),
:page 5},
:content {:text " we analyze the types of noises in this sectio"},
:properties {:color "blue"}}
{:id #uuid "6527c239-2278-4e06-bbbf-77bc08c30ddf",
:page 5,
:position {:bounding {:x1 0,
:y1 385.75213623046875,
:x2 1077.5132293701172,
:y2 977.5991516113281,
:width 1224,
:height 1584},
:rects ({:x1 1056.729965209961,
:y1 385.75213623046875,
:x2 1077.5132293701172,
:y2 405.8352966308594,
:width 1224,
:height 1584}
{:x1 220.315673828125,
:y1 413.15966796875,
:x2 1075.70947265625,
:y2 433.2428283691406,
:width 1224,
:height 1584}
{:x1 0,
:y1 957.5159606933594,
:x2 0,
:y2 977.5991516113281,
:width 1224,
:height 1584}),
:page 5},
:content {:text "we define “noise” in the dataset as data samples where the target output cannot be generated based on the existing input"},
:properties {:color "purple"}}
{:id #uuid "6527c25b-b318-43a5-94eb-a2de19e37990",
:page 5,
:position {:bounding {:x1 0,
:y1 440.4031524658203,
:x2 1077.511245727539,
:y2 1009.5924377441406,
:width 1224,
:height 1584},
:rects ({:x1 564.8589019775391,
:y1 440.4031524658203,
:x2 1077.511245727539,
:y2 460.48631286621094,
:width 1224,
:height 1584}
{:x1 220.315673828125,
:y1 467.8035430908203,
:x2 332.3496856689453,
:y2 487.8866882324219,
:width 1224,
:height 1584}
{:x1 0,
:y1 989.5092468261719,
:x2 0,
:y2 1009.5924377441406,
:width 1224,
:height 1584}),
:page 5},
:content {:text " there is a significant semantic gap between the input and the output after an update"},
:properties {:color "purple"}}
{:id #uuid "6527c3eb-297a-44c0-99c2-34cdb14e8a9d",
:page 5,
:position {:bounding {:x1 287.0672912597656,
:y1 577.4193420410156,
:x2 842.8819274902344,
:y2 597.5025024414062,
:width 1224,
:height 1584},
:rects ({:x1 287.0672912597656,
:y1 577.4193420410156,
:x2 842.8819274902344,
:y2 597.5025024414062,
:width 1224,
:height 1584}),
:page 5},
:content {:text " Below we discuss specific situations of these two types of weak correlations"},
:properties {:color "yellow"}}
{:id #uuid "6527c42a-9e06-422f-89f9-42008f798773",
:page 5,
:position {:bounding {:x1 0,
:y1 1101.4856872558594,
:x2 1077.524642944336,
:y2 1227.7756652832031,
:width 1224,
:height 1584},
:rects ({:x1 0,
:y1 1101.4856872558594,
:x2 0,
:y2 1121.5688781738281,
:width 1224,
:height 1584}
{:x1 940.3826751708984,
:y1 1180.2850036621094,
:x2 1077.524642944336,
:y2 1200.3681945800781,
:width 1224,
:height 1584}
{:x1 220.315673828125,
:y1 1207.6925354003906,
:x2 416.57261657714844,
:y2 1227.7756652832031,
:width 1224,
:height 1584}),
:page 5},
:content {:text " invalid comments can be found among them"},
:properties {:color "green"}}
{:id #uuid "6527c63f-b3fb-4384-9a8f-db163ca17ea3",
:page 5,
:position {:bounding {:x1 0,
:y1 1133.4789733886719,
:x2 1077.5083465576172,
:y2 1282.5764770507812,
:width 1224,
:height 1584},
:rects ({:x1 0,
:y1 1133.4789733886719,
:x2 0,
:y2 1153.5621337890625,
:width 1224,
:height 1584}
{:x1 694.4877166748047,
:y1 1235.0929565429688,
:x2 1077.5083465576172,
:y2 1255.1760864257812,
:width 1224,
:height 1584}
{:x1 220.315673828125,
:y1 1262.4933471679688,
:x2 920.0357666015625,
:y2 1282.5764770507812,
:width 1224,
:height 1584}),
:page 5},
:content {:text "comments that consist of purely invalid characters and comments that only contain a single unrelated word do not fit the scenario of real updates"},
:properties {:color "green"}}
{:id #uuid "6527c660-e045-4759-9a79-a2d6c9bee964",
:page 6,
:position {:bounding {:x1 625.4665451049805,
:y1 221.33551025390625,
:x2 851.7538070678711,
:y2 241.418701171875,
:width 1224,
:height 1584},
:rects ({:x1 625.4665451049805,
:y1 221.33551025390625,
:x2 851.7538070678711,
:y2 241.418701171875,
:width 1224,
:height 1584}),
:page 6},
:content {:text "new code has been deprecated"},
:properties {:color "blue"}}
{:id #uuid "6527c67f-a00e-4fda-a819-e76337de3bec",
:page 6,
:position {:bounding {:x1 0,
:y1 330.95135498046875,
:x2 1004.0995407104492,
:y2 961.6024780273438,
:width 1224,
:height 1584},
:rects ({:x1 329.2383346557617,
:y1 330.95135498046875,
:x2 1004.0995407104492,
:y2 351.03448486328125,
:width 1224,
:height 1584}
{:x1 146.26616668701172,
:y1 358.35174560546875,
:x2 189.26712799072266,
:y2 378.43487548828125,
:width 1224,
:height 1584}
{:x1 0,
:y1 941.5193481445312,
:x2 0,
:y2 961.6024780273438,
:width 1224,
:height 1584}),
:page 6},
:content {:text " correlation between the code changes and the corresponding comment updating may also be weak."},
:properties {:color "green"}}
{:id #uuid "6527c68a-1ca0-4ee1-9a25-a8ca3da895c9",
:page 6,
:position {:bounding {:x1 279.51615142822266,
:y1 358.35174560546875,
:x2 1002.3132095336914,
:y2 378.43487548828125,
:width 1224,
:height 1584},
:rects ({:x1 279.51615142822266,
:y1 358.35174560546875,
:x2 1002.3132095336914,
:y2 378.43487548828125,
:width 1224,
:height 1584}),
:page 6},
:content {:text " if the code is significantly modified but the comments barely change, it belongs to this type of noise"},
:properties {:color "blue"}}
{:id #uuid "6527c6c6-b750-4172-85cd-908563a8440c",
:page 6,
:position {:bounding {:x1 0,
:y1 413.1596374511719,
:x2 1004.1308517456055,
:y2 1009.5924377441406,
:width 1224,
:height 1584},
:rects ({:x1 183.79027557373047,
:y1 413.1596374511719,
:x2 1004.1308517456055,
:y2 433.2428283691406,
:width 1224,
:height 1584}
{:x1 146.87950134277344,
:y1 440.5600280761719,
:x2 222.9415283203125,
:y2 460.6432189941406,
:width 1224,
:height 1584}
{:x1 0,
:y1 989.5093078613281,
:x2 0,
:y2 1009.5924377441406,
:width 1224,
:height 1584}),
:page 6},
:content {:text "Similarly, if the code barely changes while the comments need to be significantly modified, this can also mislead the model."},
:properties {:color "blue"}}
{:id #uuid "6527c6e3-a425-4038-b5d0-85f2b49eac3d",
:page 6,
:position {:bounding {:x1 0,
:y1 440.5600280761719,
:x2 1003.1842575073242,
:y2 1025.5890197753906,
:width 1224,
:height 1584},
:rects ({:x1 326.7666549682617,
:y1 440.5600280761719,
:x2 1003.1842575073242,
:y2 460.6432189941406,
:width 1224,
:height 1584}
{:x1 146.87950134277344,
:y1 467.9675598144531,
:x2 334.3770294189453,
:y2 488.0507507324219,
:width 1224,
:height 1584}
{:x1 0,
:y1 1005.5058898925781,
:x2 0,
:y2 1025.5890197753906,
:width 1224,
:height 1584}),
:page 6},
:content {:text "oise can exist in data where both comments and code undergo changes but their changes are not related to each other. "},
:properties {:color "blue"}}
{:id #uuid "6527c751-fdb2-4348-8cb2-77a8bb9ce1b2",
:page 6,
:position {:bounding {:x1 0,
:y1 495.3679428100586,
:x2 1004.1022415161133,
:y2 1089.575584411621,
:width 1224,
:height 1584},
:rects ({:x1 166.82715606689453,
:y1 495.3679428100586,
:x2 1004.1022415161133,
:y2 515.4511032104492,
:width 1224,
:height 1584}
{:x1 146.87950134277344,
:y1 522.7683334350586,
:x2 1003.7802581787109,
:y2 542.8514938354492,
:width 1224,
:height 1584}
{:x1 146.87950134277344,
:y1 550.1758651733398,
:x2 1003.7744598388672,
:y2 570.2590255737305,
:width 1224,
:height 1584}
{:x1 146.87950134277344,
:y1 577.5762557983398,
:x2 820.7284393310547,
:y2 597.6594161987305,
:width 1224,
:height 1584}
{:x1 0,
:y1 1037.499168395996,
:x2 0,
:y2 1057.5822982788086,
:width 1224,
:height 1584}
{:x1 0,
:y1 1053.4958114624023,
:x2 0,
:y2 1073.5789413452148,
:width 1224,
:height 1584}
{:x1 0,
:y1 1069.4923934936523,
:x2 0,
:y2 1089.575584411621,
:width 1224,
:height 1584}),
:page 6},
:content {:text "In summary, weak correlations between comments and code, as well as weak correlations between comment changes and code changes, are typical reasons for data noise in datasets for comment updating. These noisy data can mislead the model’s generation process, thus reducing the overall performance. Therefore, we may need a criteria to evaluate the quality of individual data samples and filter out noises based on their data quality scores"},
:properties {:color "purple"}}
{:id #uuid "6527c78e-33c1-4cce-ae58-79cf5b88a983",
:page 6,
:position {:bounding {:x1 362.06006622314453,
:y1 745.7941284179688,
:x2 612.9912338256836,
:y2 765.8772583007812,
:width 1224,
:height 1584},
:rects ({:x1 362.06006622314453,
:y1 745.7941284179688,
:x2 612.9912338256836,
:y2 765.8772583007812,
:width 1224,
:height 1584}),
:page 6},
:content {:text " anchor to filter out the noising data"},
:properties {:color "yellow"}}
{:id #uuid "6527ea08-470e-48cd-a341-246bfa44dbf5",
:page 6,
:position {:bounding {:x1 0,
:y1 745.0595703125,
:x2 870.2991943359375,
:y2 1233.5453491210938,
:width 1060.8,
:height 1372.8},
:rects ({:x1 633.2590942382812,
:y1 745.0595703125,
:x2 870.2991943359375,
:y2 762.4041137695312,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 768.8084716796875,
:x2 371.16259765625,
:y2 786.1530151367188,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1213.462158203125,
:x2 0,
:y2 1233.5453491210938,
:width 1060.8,
:height 1372.8}),
:page 6},
:content {:text " we designed a criterion to calculate the quality scores for each data samp"},
:properties {:color "blue"}}
{:id #uuid "6527ea11-4785-4a7b-acc7-ed57e72e64bb",
:page 6,
:position {:bounding {:x1 562.720458984375,
:y1 792.557373046875,
:x2 737.4953002929688,
:y2 809.9019165039062,
:width 1060.8,
:height 1372.8},
:rects ({:x1 562.720458984375,
:y1 792.557373046875,
:x2 737.4953002929688,
:y2 809.9019165039062,
:width 1060.8,
:height 1372.8}),
:page 6},
:content {:text " internal semantic similarity"},
:properties {:color "purple"}}
{:id #uuid "6527ea21-7aca-4a90-9625-8689f2281167",
:page 6,
:position {:bounding {:x1 144.5830535888672,
:y1 879.0447387695312,
:x2 537.5011444091797,
:y2 896.3893127441406,
:width 1060.8,
:height 1372.8},
:rects ({:x1 144.5830535888672,
:y1 879.0447387695312,
:x2 537.5011444091797,
:y2 896.3893127441406,
:width 1060.8,
:height 1372.8}),
:page 6},
:content {:text "Weak correlations existing between comments and code (Type I)"},
:properties {:color "green"}}
{:id #uuid "6527ea2b-87d4-411c-9f87-6ea17e5dff20",
:page 6,
:position {:bounding {:x1 655.6843872070312,
:y1 902.7936706542969,
:x2 845.3291625976562,
:y2 920.1382141113281,
:width 1060.8,
:height 1372.8},
:rects ({:x1 655.6843872070312,
:y1 902.7936706542969,
:x2 845.3291625976562,
:y2 920.1382141113281,
:width 1060.8,
:height 1372.8}),
:page 6},
:content {:text "r internal semantic similarity"},
:properties {:color "yellow"}}
{:id #uuid "6527eb2c-3d1e-496a-95fc-f0bd677b27e4",
:page 6,
:position {:bounding {:x1 0,
:y1 997.7892761230469,
:x2 870.2553100585938,
:y2 1393.5116882324219,
:width 1060.8,
:height 1372.8},
:rects ({:x1 420.1414794921875,
:y1 997.7892761230469,
:x2 870.2553100585938,
:y2 1015.1338195800781,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 1021.4026794433594,
:x2 219.98709106445312,
:y2 1038.7472229003906,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1373.4284973144531,
:x2 0,
:y2 1393.5116882324219,
:width 1060.8,
:height 1372.8}),
:page 6},
:content {:text ". It is worth noting that the semantic representation mentioned above is at the token level."},
:properties {:color "blue"}}
{:id #uuid "6527eb30-56b8-4463-8dca-b866af730ee3",
:page 6,
:position {:bounding {:x1 0,
:y1 974.0403747558594,
:x2 871.5245971679688,
:y2 1377.5150451660156,
:width 1060.8,
:height 1372.8},
:rects ({:x1 291.9307556152344,
:y1 974.0403747558594,
:x2 871.5245971679688,
:y2 991.3849182128906,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 997.7892761230469,
:x2 412.4386901855469,
:y2 1015.1338195800781,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1357.4319152832031,
:x2 0,
:y2 1377.5150451660156,
:width 1060.8,
:height 1372.8}),
:page 6},
:content {:text "We choose the state-of-the-art model pre-trained on code-related datasets, GraphCodeBert [ 8 ], to calculate the semantics of comments or cod"},
:properties {:color "green"}}
{:id #uuid "6527eb4f-5388-42aa-9b20-6252653d43f0",
:page 6,
:position {:bounding {:x1 0,
:y1 1045.151611328125,
:x2 870.147705078125,
:y2 1425.5049438476562,
:width 1060.8,
:height 1372.8},
:rects ({:x1 608.8800048828125,
:y1 1045.151611328125,
:x2 870.147705078125,
:y2 1062.4961547851562,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 1068.9005126953125,
:x2 812.1407775878906,
:y2 1086.2450561523438,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1405.4218139648438,
:x2 0,
:y2 1425.5049438476562,
:width 1060.8,
:height 1372.8}),
:page 6},
:content {:text " cosine similarity of semantic embeddings to describe the correlation between old and new comments, and between old and new code, respectively"},
:properties {:color "purple"}}
{:id #uuid "6527ebde-d308-4bfa-b756-f1357554b3e2",
:page 7,
:position {:bounding {:x1 0,
:y1 444.6893615722656,
:x2 933.9067077636719,
:y2 1265.5386047363281,
:width 1060.8,
:height 1372.8},
:rects ({:x1 609.5851440429688,
:y1 444.6893615722656,
:x2 933.4479370117188,
:y2 462.0339050292969,
:width 1060.8,
:height 1372.8}
{:x1 190.93978881835938,
:y1 468.4382629394531,
:x2 933.9067077636719,
:y2 485.7828063964844,
:width 1060.8,
:height 1372.8}
{:x1 190.93978881835938,
:y1 492.1943054199219,
:x2 304.92803955078125,
:y2 509.5388488769531,
:width 1060.8,
:height 1372.8}
{:x1 316.217041015625,
:y1 492.1943054199219,
:x2 502.00372314453125,
:y2 509.5388488769531,
:width 1060.8,
:height 1372.8}
{:x1 305.612060546875,
:y1 497.9568176269531,
:x2 319.0127258300781,
:y2 512.1062927246094,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1229.4587707519531,
:x2 0,
:y2 1249.5419006347656,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1245.4554748535156,
:x2 0,
:y2 1265.5386047363281,
:width 1060.8,
:height 1372.8}),
:page 7},
:content {:text "Sentence-BERT helps in capturing natural language semantic information to enhance the overall performance and robustness of the evaluation. Finally, we compute the evaluation score 𝑆1 for identifying noise in type I"},
:properties {:color "purple"}}
{:id #uuid "6527ec07-8c4d-491c-bd5d-abed4cf3c0cd",
:page 7,
:position {:bounding {:x1 208.23440551757812,
:y1 536.5326995849609,
:x2 700.0862426757812,
:y2 553.8773040771484,
:width 1060.8,
:height 1372.8},
:rects ({:x1 208.23440551757812,
:y1 536.5326995849609,
:x2 700.0862426757812,
:y2 553.8773040771484,
:width 1060.8,
:height 1372.8}),
:page 7},
:content {:text "Weak correlation existing between comment changes and code changes (Type II)"},
:properties {:color "green"}}
{:id #uuid "6527ec5d-8a25-4040-9011-35f017bda998",
:page 7,
:position {:bounding {:x1 455.7780456542969,
:y1 560.2816314697266,
:x2 853.10498046875,
:y2 577.6261749267578,
:width 1060.8,
:height 1372.8},
:rects ({:x1 455.7780456542969,
:y1 560.2816314697266,
:x2 853.10498046875,
:y2 577.6261749267578,
:width 1060.8,
:height 1372.8}),
:page 7},
:content {:text "nconsistencies between comment changes and code changes"},
:properties {:color "blue"}}
{:id #uuid "6527ed40-630d-44d2-86be-06711677011d",
:page 9,
:position {:bounding {:x1 0,
:y1 828.6656799316406,
:x2 933.63134765625,
:y2 1361.5183410644531,
:width 1060.8,
:height 1372.8},
:rects ({:x1 872.6842041015625,
:y1 828.6656799316406,
:x2 933.63134765625,
:y2 846.0102233886719,
:width 1060.8,
:height 1372.8}
{:x1 190.93978881835938,
:y1 852.4145812988281,
:x2 276.651123046875,
:y2 869.7591247558594,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1341.4352111816406,
:x2 0,
:y2 1361.5183410644531,
:width 1060.8,
:height 1372.8}),
:page 9},
:content {:text "comment updating task"},
:properties {:color "blue"}}
{:id #uuid "6527ed50-1f6f-4ad9-be71-de06a3c61a27",
:page 9,
:position {:bounding {:x1 609.5426635742188,
:y1 852.4145812988281,
:x2 716.8668823242188,
:y2 869.7591247558594,
:width 1060.8,
:height 1372.8},
:rects ({:x1 609.5426635742188,
:y1 852.4145812988281,
:x2 716.8668823242188,
:y2 869.7591247558594,
:width 1060.8,
:height 1372.8}),
:page 9},
:content {:text "pdated comment"},
:properties {:color "blue"}}
{:id #uuid "65280a83-cf57-4801-b278-342075aba5b8",
:page 9,
:position {:bounding {:x1 304.2029113769531,
:y1 779.3706665039062,
:x2 531.6401977539062,
:y2 795.3458862304688,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 304.2029113769531,
:y1 779.3706665039062,
:x2 531.6401977539062,
:y2 795.3458862304688,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text "metrics for evaluating the effectiveness"},
:properties {:color "blue"}}
{:id #uuid "65280a8a-682c-4fe4-a495-96eb196316c4",
:page 9,
:position {:bounding {:x1 772.4931640625,
:y1 779.3706665039062,
:x2 831.8060302734375,
:y2 795.3458862304688,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 772.4931640625,
:y1 779.3706665039062,
:x2 831.8060302734375,
:y2 795.3458862304688,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text "baselines "},
:properties {:color "blue"}}
{:id #uuid "65280a8f-37f8-4c01-b348-8596dc431019",
:page 9,
:position {:bounding {:x1 174.59368896484375,
:y1 801.0869750976562,
:x2 288.6856994628906,
:y2 817.062255859375,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 174.59368896484375,
:y1 801.0869750976562,
:x2 288.6856994628906,
:y2 817.062255859375,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text "experiment settings"},
:properties {:color "blue"}}
{:id #uuid "65280a9e-80e8-4c23-82b7-0b3ab47c7501",
:page 9,
:position {:bounding {:x1 223.46270751953125,
:y1 901.1319274902344,
:x2 274.8182067871094,
:y2 917.1072082519531,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 223.46270751953125,
:y1 901.1319274902344,
:x2 274.8182067871094,
:y2 917.1072082519531,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text "Whether "},
:properties {:color "red"}}
{:id #uuid "65280b47-e7d9-4af7-84b0-3050101288de",
:page 9,
:position {:bounding {:x1 264.5247802734375,
:y1 922.8482971191406,
:x2 446.7215576171875,
:y2 938.8235168457031,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 264.5247802734375,
:y1 922.8482971191406,
:x2 446.7215576171875,
:y2 938.8235168457031,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text "changes in the number of data"},
:properties {:color "green"}}
{:id #uuid "65280bb0-8987-4b66-9bd8-f3d4821482a2",
:page 9,
:position {:bounding {:x1 0,
:y1 922.8482971191406,
:x2 853.5363159179688,
:y2 1457.4982604980469,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 564.9109497070312,
:y1 922.8482971191406,
:x2 853.5363159179688,
:y2 938.8235168457031,
:width 970,
:height 1255.2941176470588}
{:x1 174.59368896484375,
:y1 944.5646057128906,
:x2 274.77825927734375,
:y2 960.5398864746094,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1437.4150085449219,
:x2 0,
:y2 1457.4982604980469,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text "xamples that CupCleaner identifies as noisy and high-quality data,"},
:properties {:color "blue"}}
{:id #uuid "65280bb8-4b66-40e2-a189-ca3d373cdd49",
:page 9,
:position {:bounding {:x1 0,
:y1 944.5646057128906,
:x2 854.1444702148438,
:y2 1473.4948425292969,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 472.4082336425781,
:y1 944.5646057128906,
:x2 854.1444702148438,
:y2 960.5398864746094,
:width 970,
:height 1255.2941176470588}
{:x1 174.59368896484375,
:y1 966.2881164550781,
:x2 240.87661743164062,
:y2 982.2633361816406,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1453.4117126464844,
:x2 0,
:y2 1473.4948425292969,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text " human evaluation on the noisy and high-quality data identified by CupCleane"},
:properties {:color "blue"}}
{:id #uuid "65280bc1-dd33-4b4c-b521-b847608eeb33",
:page 9,
:position {:bounding {:x1 199.69802856445312,
:y1 1000.050048828125,
:x2 555.04541015625,
:y2 1016.0252685546875,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 199.69802856445312,
:y1 1000.050048828125,
:x2 555.04541015625,
:y2 1016.0252685546875,
:width 970,
:height 1255.2941176470588}),
:page 9},
:content {:text "Q2: How effective is the training data cleaned by CupCleaner?"},
:properties {:color "green"}}
{:id #uuid "65280bf7-8a02-4e7f-bb1e-dce324699404",
:page 10,
:position {:bounding {:x1 441.7239990234375,
:y1 375.796142578125,
:x2 769.2066650390625,
:y2 391.77142333984375,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 441.7239990234375,
:y1 375.796142578125,
:x2 769.2066650390625,
:y2 391.77142333984375,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text " CupCleaner to clean only the training and validation sets"},
:properties {:color "green"}}
{:id #uuid "65280c01-34a8-48c7-9f34-86e59c418846",
:page 10,
:position {:bounding {:x1 0,
:y1 397.51251220703125,
:x2 795.587158203125,
:y2 1009.5924682617188,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 375.46466064453125,
:y1 397.51251220703125,
:x2 795.587158203125,
:y2 413.48773193359375,
:width 970,
:height 1255.2941176470588}
{:x1 115.9132080078125,
:y1 419.22882080078125,
:x2 391.3471984863281,
:y2 435.20404052734375,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 989.5092163085938,
:x2 0,
:y2 1009.5924682617188,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "comparing the performance of the models trained on the cleaned dataset with the models trained on the original dataset "},
:properties {:color "blue"}}
{:id #uuid "65280c1c-24d1-4a81-bc17-dbec3d6ab178",
:page 10,
:position {:bounding {:x1 132.20938110351562,
:y1 452.86956787109375,
:x2 572.6553344726562,
:y2 468.84478759765625,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 132.20938110351562,
:y1 452.86956787109375,
:x2 572.6553344726562,
:y2 468.84478759765625,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "RQ3: How effective is the use of CupCleaner-cleaned data for evaluating models? "},
:properties {:color "yellow"}}
{:id #uuid "65281639-713e-4ff7-921e-6fb204174b3b",
:page 10,
:position {:bounding {:x1 0,
:y1 474.5858917236328,
:x2 795.584228515625,
:y2 1057.582290649414,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 696.0598754882812,
:y1 474.5858917236328,
:x2 795.584228515625,
:y2 490.5611114501953,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 496.30223083496094,
:x2 617.7791213989258,
:y2 512.2774810791016,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1037.4991607666016,
:x2 0,
:y2 1057.582290649414,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "different versions of the test set using the best-performing model from the previous research question. "},
:properties {:color "green"}}
{:id #uuid "6528166d-bbb6-4570-9178-041617f45ebb",
:page 10,
:position {:bounding {:x1 132.20938110351562,
:y1 573.3756256103516,
:x2 329.4228210449219,
:y2 589.3508758544922,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 132.20938110351562,
:y1 573.3756256103516,
:x2 329.4228210449219,
:y2 589.3508758544922,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "RQ4: How efficient is CupCleaner?"},
:properties {:color "green"}}
{:id #uuid "65281694-0f23-4f97-bb44-691e9c91a6e7",
:page 10,
:position {:bounding {:x1 0,
:y1 616.8154449462891,
:x2 796.9031372070312,
:y2 1153.5620880126953,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 402.5033874511719,
:y1 616.8154449462891,
:x2 796.9031372070312,
:y2 632.7906646728516,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 638.5318145751953,
:x2 516.2017288208008,
:y2 654.5070343017578,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1133.4789581298828,
:x2 0,
:y2 1153.5620880126953,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "we calculate the time consumption for cleaning the entire dataset, including the time required to compute scores and filter out the data"},
:properties {:color "green"}}
{:id #uuid "65281b6d-8201-421e-be12-794ebdc336a8",
:page 10,
:position {:bounding {:x1 0,
:y1 748.1119728088379,
:x2 795.2285766601562,
:y2 1233.5452613830566,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 582.37744140625,
:y1 748.1119728088379,
:x2 795.2285766601562,
:y2 764.0871925354004,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 769.8282814025879,
:x2 234.6419677734375,
:y2 785.8035621643066,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1213.4621315002441,
:x2 0,
:y2 1233.5452613830566,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "able 1 describes the basic statistics of these three datasets."},
:properties {:color "green"}}
{:id #uuid "65281bb4-a458-45d0-b6a1-afd7d6980e06",
:page 10,
:position {:bounding {:x1 0,
:y1 834.9844512939453,
:x2 795.5413208007812,
:y2 1297.5318145751953,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 225.2294921875,
:y1 834.9844512939453,
:x2 795.5413208007812,
:y2 850.9596710205078,
:width 970,
:height 1255.2941176470588}
{:x1 116.19847869873047,
:y1 856.7008209228516,
:x2 252.27434539794922,
:y2 872.6760406494141,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1277.4486846923828,
:x2 0,
:y2 1297.5318145751953,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "Panthaplackel et al., 2020 dataset collects 7.2k data samples from the commit history of open-source Java projects on Github"},
:properties {:color "green"}}
{:id #uuid "65281bbd-0377-41e6-85cc-b25f16325046",
:page 10,
:position {:bounding {:x1 0,
:y1 856.7008209228516,
:x2 795.48583984375,
:y2 1313.5285186767578,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 285.27215576171875,
:y1 856.7008209228516,
:x2 795.48583984375,
:y2 872.6760406494141,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 878.4171295166016,
:x2 348.47757720947266,
:y2 894.3924102783203,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1293.4453887939453,
:x2 0,
:y2 1313.5285186767578,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "Liu et al., 2020 dataset is constructed from 1,063K method-doc co-change instances from Github, resulting in 104k data samples"},
:properties {:color "green"}}
{:id #uuid "65281bc7-fd1d-43f4-ac76-4d9be7d8299f",
:page 10,
:position {:bounding {:x1 0,
:y1 878.4171295166016,
:x2 795.4835205078125,
:y2 1329.5251007080078,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 384.211669921875,
:y1 878.4171295166016,
:x2 795.4835205078125,
:y2 894.3924102783203,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 900.1334991455078,
:x2 648.6476211547852,
:y2 916.1087188720703,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1309.4419708251953,
:x2 0,
:y2 1329.5251007080078,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "Panthaplackel et al., 2021 dataset is originally designed for comment consistency detection, but some of the data can also be used for the comment updating task"},
:properties {:color "green"}}
{:id #uuid "65281c21-7981-447a-9611-a880f5781076",
:page 10,
:position {:bounding {:x1 0,
:y1 900.1334991455078,
:x2 795.5892333984375,
:y2 1361.5183868408203,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 655.88671875,
:y1 900.1334991455078,
:x2 795.5892333984375,
:y2 916.1087188720703,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 921.8498077392578,
:x2 795.5811538696289,
:y2 937.8250885009766,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 943.5661773681641,
:x2 217.99407958984375,
:y2 959.5413970947266,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1325.4385528564453,
:x2 0,
:y2 1345.5218048095703,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1341.4352569580078,
:x2 0,
:y2 1361.5183868408203,
:width 970,
:height 1255.2941176470588}),
:page 10},
:content {:text "We extract the data with comment changes from this dataset to construct a new comment updating dataset, and discard the data with no changes in the comments. "},
:properties {:color "yellow"}}
{:id #uuid "65281c6a-e936-4cd6-aac0-d7fc79cf8632",
:page 11,
:position {:bounding {:x1 0,
:y1 240.72686767578125,
:x2 853.7792358398438,
:y2 945.6058349609375,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 174.59368896484375,
:y1 240.72686767578125,
:x2 853.7792358398438,
:y2 256.7021484375,
:width 970,
:height 1255.2941176470588}
{:x1 174.59368896484375,
:y1 262.4432373046875,
:x2 371.5769958496094,
:y2 278.41845703125,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 925.522705078125,
:x2 0,
:y2 945.6058349609375,
:width 970,
:height 1255.2941176470588}),
:page 11},
:content {:text "However, due to a lack of exploration into the characteristics of comment updating, the constructed datasets still contain a significant amount of noisy data"},
:properties {:color "yellow"}}
{:id #uuid "65281cb8-75c4-4555-b8d1-aae95c63a161",
:page 11,
:position {:bounding {:x1 438.05279541015625,
:y1 351.8190002441406,
:x2 499.6300048828125,
:y2 367.79425048828125,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 438.05279541015625,
:y1 351.8190002441406,
:x2 499.6300048828125,
:y2 367.79425048828125,
:width 970,
:height 1255.2941176470588}),
:page 11},
:content {:text "oken level "},
:properties {:color "green"}}
{:id #uuid "65281cbc-9398-4989-bbf8-bf86684392ae",
:page 11,
:position {:bounding {:x1 630.603515625,
:y1 351.8190002441406,
:x2 727.2633666992188,
:y2 367.79425048828125,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 630.603515625,
:y1 351.8190002441406,
:x2 727.2633666992188,
:y2 367.79425048828125,
:width 970,
:height 1255.2941176470588}),
:page 11},
:content {:text "e sentence level "},
:properties {:color "green"}}
{:id #uuid "65281cc5-b0a4-4d47-b42f-bd0d44565f40",
:page 11,
:position {:bounding {:x1 582.1118774414062,
:y1 330.1026802062988,
:x2 853.8264770507812,
:y2 346.0778999328613,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 582.1118774414062,
:y1 330.1026802062988,
:x2 853.8264770507812,
:y2 346.0778999328613,
:width 970,
:height 1255.2941176470588}),
:page 11},
:content {:text " evaluation metrics in the field of text generation"},
:properties {:color "green"}}
{:id #uuid "65281d03-bc13-4652-aaab-197042fe3e1e",
:page 11,
:position {:bounding {:x1 697.0545654296875,
:y1 666.0177803039551,
:x2 729.2894287109375,
:y2 681.9930000305176,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 697.0545654296875,
:y1 666.0177803039551,
:x2 729.2894287109375,
:y2 681.9930000305176,
:width 970,
:height 1255.2941176470588}),
:page 11},
:content {:text "align "},
:properties {:color "yellow"}}
{:id #uuid "65281d43-dd7a-4892-b487-4dcc5d352db9",
:page 11,
:position {:bounding {:x1 0,
:y1 744.6031074523926,
:x2 852.816650390625,
:y2 1249.5418891906738,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 676.40234375,
:y1 744.6031074523926,
:x2 852.816650390625,
:y2 760.5783271789551,
:width 970,
:height 1255.2941176470588}
{:x1 174.59368896484375,
:y1 766.3194770812988,
:x2 629.5917358398438,
:y2 782.2946968078613,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1229.4587593078613,
:x2 0,
:y2 1249.5418891906738,
:width 970,
:height 1255.2941176470588}),
:page 11},
:content {:text "valuation metric based on edit distance, originally designed to assess the overall quality of text simplification."},
:properties {:color "green"}}
{:id #uuid "65281d78-6b37-462f-b679-f28724513791",
:page 11,
:position {:bounding {:x1 0,
:y1 888.2091064453125,
:x2 853.77392578125,
:y2 1345.521728515625,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 366.9482421875,
:y1 888.2091064453125,
:x2 853.77392578125,
:y2 904.184326171875,
:width 970,
:height 1255.2941176470588}
{:x1 174.59368896484375,
:y1 909.9254760742188,
:x2 260.18505859375,
:y2 925.9006958007812,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1325.4385986328125,
:x2 0,
:y2 1345.521728515625,
:width 970,
:height 1255.2941176470588}),
:page 11},
:content {:text "GLEU is closer to human-level judgment than BLEU and is more suitable for tasks involving edits"},
:properties {:color "green"}}
{:id #uuid "65281df5-34e7-4335-80e7-73c907dc3802",
:page 12,
:position {:bounding {:x1 155.58743286132812,
:y1 213.112548828125,
:x2 201.92930603027344,
:y2 229.08779907226562,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 155.58743286132812,
:y1 213.112548828125,
:x2 201.92930603027344,
:y2 229.08779907226562,
:width 970,
:height 1255.2941176470588}),
:page 12},
:content {:text "PLBART"},
:properties {:color "purple"}}
{:id #uuid "65281e15-0e25-43ed-bded-06f26fb91853",
:page 12,
:position {:bounding {:x1 0,
:y1 446.21556091308594,
:x2 795.1854248046875,
:y2 1073.5789642333984,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 596.2813110351562,
:y1 446.21556091308594,
:x2 795.1854248046875,
:y2 462.19078063964844,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 467.81065368652344,
:x2 499.8027114868164,
:y2 483.7859344482422,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1053.495834350586,
:x2 0,
:y2 1073.5789642333984,
:width 970,
:height 1255.2941176470588}),
:page 12},
:content {:text "he experimental results show that UniXcoder achieves improvements on multiple code-related tasks."},
:properties {:color "purple"}}
{:id #uuid "65281e1e-0235-4784-810a-60d8435d3d0a",
:page 12,
:position {:bounding {:x1 0,
:y1 278.26158142089844,
:x2 795.4754638671875,
:y2 977.5991058349609,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 652.1166381835938,
:y1 278.26158142089844,
:x2 795.357421875,
:y2 294.23683166503906,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 299.97792053222656,
:x2 795.4754638671875,
:y2 315.9531707763672,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 321.6942596435547,
:x2 211.36370849609375,
:y2 337.6695098876953,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 941.5193939208984,
:x2 0,
:y2 961.6025238037109,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 957.5159759521484,
:x2 0,
:y2 977.5991058349609,
:width 970,
:height 1255.2941176470588}),
:page 12},
:content {:text "The experimental results demonstrate that PLBART outperforms CodeGPT[ 18 ] and CodeBERT[ 5 ] in multiple code understanding and code generation tasks"},
:properties {:color "purple"}}
{:id #uuid "65281e56-7510-44fc-9b22-53b9e89d5c9d",
:page 12,
:position {:bounding {:x1 590.9583740234375,
:y1 705.8132171630859,
:x2 665.6791381835938,
:y2 721.7884979248047,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 590.9583740234375,
:y1 705.8132171630859,
:x2 665.6791381835938,
:y2 721.7884979248047,
:width 970,
:height 1255.2941176470588}),
:page 12},
:content {:text "188GB RAM"},
:properties {:color "purple"}}
{:id #uuid "65281e5d-e30b-4338-abfe-6c0ed7002216",
:page 12,
:position {:bounding {:x1 0,
:y1 705.8132171630859,
:x2 794.6439819335938,
:y2 1233.5453338623047,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 706.3788452148438,
:y1 705.8132171630859,
:x2 794.6439819335938,
:y2 721.7884979248047,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 727.5295867919922,
:x2 313.5955581665039,
:y2 743.5048065185547,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1213.4622039794922,
:x2 0,
:y2 1233.5453338623047,
:width 970,
:height 1255.2941176470588}),
:page 12},
:content {:text " NVIDIA TITAN RTX GPU with 24GB of memory."},
:properties {:color "purple"}}
{:id #uuid "65281e63-04fc-48df-934a-b1b687e0951b",
:page 12,
:position {:bounding {:x1 179.83399963378906,
:y1 749.2458953857422,
:x2 194.95497131347656,
:y2 765.2211761474609,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 179.83399963378906,
:y1 749.2458953857422,
:x2 194.95497131347656,
:y2 765.2211761474609,
:width 970,
:height 1255.2941176470588}),
:page 12},
:content {:text "20"},
:properties {:color "purple"}}
{:id #uuid "65281e6f-e9ea-4ab6-9352-51c962ce0602",
:page 12,
:position {:bounding {:x1 0,
:y1 770.9622650146484,
:x2 795.58251953125,
:y2 1281.5352020263672,
:width 970,
:height 1255.2941176470588},
:rects ({:x1 277.535888671875,
:y1 770.9622650146484,
:x2 795.58251953125,
:y2 786.9374847412109,
:width 970,
:height 1255.2941176470588}
{:x1 116.3981704711914,
:y1 792.6785736083984,
:x2 385.62967681884766,
:y2 808.6538543701172,
:width 970,
:height 1255.2941176470588}
{:x1 0,
:y1 1261.4520721435547,
:x2 0,
:y2 1281.5352020263672,
:width 970,
:height 1255.2941176470588}),
:page 12},
:content {:text "terminate the training process if the experimental performance on the validation set does not improve within three consecutive epochs. F"},
:properties {:color "purple"}}
{:id #uuid "65282097-c213-4409-93a4-fcec00a651ad",
:page 13,
:position {:bounding {:x1 0,
:y1 785.4183959960938,
:x2 933.6729583740234,
:y2 1185.555419921875,
:width 1060.8,
:height 1372.8},
:rects ({:x1 395.0675506591797,
:y1 785.4183959960938,
:x2 933.6729583740234,
:y2 802.762939453125,
:width 1060.8,
:height 1372.8}
{:x1 190.93977546691895,
:y1 809.1673278808594,
:x2 636.2563953399658,
:y2 826.5118408203125,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1165.4722290039062,
:x2 0,
:y2 1185.555419921875,
:width 1060.8,
:height 1372.8}),
:page 13},
:content {:text " The red dashed line in the second step of Figure 2 (Sec.3) corresponds approximately to0.65, and the green solid line corresponds to 0.8 in the score distributio"},
:properties {:color "yellow"}}
{:id #uuid "65282108-3ce6-4d81-9758-39974f55389a",
:page 13,
:position {:bounding {:x1 208.23440551757812,
:y1 761.6694869995117,
:x2 306.3742904663086,
:y2 779.014030456543,
:width 1060.8,
:height 1372.8},
:rects ({:x1 208.23440551757812,
:y1 761.6694869995117,
:x2 306.3742904663086,
:y2 779.014030456543,
:width 1060.8,
:height 1372.8}),
:page 13},
:content {:text "The final scores"},
:properties {:color "yellow"}}
{:id #uuid "65282238-75a8-43c9-87b5-01a0f937dfcb",
:page 13,
:position {:bounding {:x1 190.41202545166016,
:y1 975.4168090820312,
:x2 908.5377578735352,
:y2 992.7613525390625,
:width 1060.8,
:height 1372.8},
:rects ({:x1 190.41202545166016,
:y1 975.4168090820312,
:x2 908.5377578735352,
:y2 992.7613525390625,
:width 1060.8,
:height 1372.8}),
:page 13},
:content {:text "As shown in Table 2, we filter out approximately 33%, 12%, and 18% of the data in the three datasets, respectively."},
:properties {:color "yellow"}}
{:id #uuid "652822d1-4eea-4c87-bc4b-dc3d3c9c1fd4",
:page 14,
:position {:bounding {:x1 0,
:y1 918.3053436279297,
:x2 870.2547225952148,
:y2 1009.5923919677734,
:width 1060.8,
:height 1372.8},
:rects ({:x1 460.07051849365234,
:y1 918.3053436279297,
:x2 870.2547225952148,
:y2 935.6498870849609,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 942.0542449951172,
:x2 186.67699432373047,
:y2 959.3987884521484,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 989.5092620849609,
:x2 0,
:y2 1009.5923919677734,
:width 1060.8,
:height 1372.8}),
:page 14},
:content {:text "“Would you accept this data sample into the comment updating dataset?\""},
:properties {:color "green"}}
{:id #uuid "652822fe-ebd6-4ef4-8eea-41a38feec147",
:page 14,
:position {:bounding {:x1 395.74340057373047,
:y1 965.6676483154297,
:x2 707.2164535522461,
:y2 983.0121917724609,
:width 1060.8,
:height 1372.8},
:rects ({:x1 395.74340057373047,
:y1 965.6676483154297,
:x2 707.2164535522461,
:y2 983.0121917724609,
:width 1060.8,
:height 1372.8}),
:page 14},
:content {:text " The label for each data sample ranges from 1 to 5"},
:properties {:color "green"}}
{:id #uuid "6528232d-b8f4-4993-983a-d3aa4f8e0374",
:page 14,
:position {:bounding {:x1 0,
:y1 1060.6632537841797,
:x2 870.4001083374023,
:y2 1125.505599975586,
:width 1060.8,
:height 1372.8},
:rects ({:x1 507.04540252685547,
:y1 1060.6632537841797,
:x2 870.4001083374023,
:y2 1078.007797241211,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 1084.4121551513672,
:x2 870.2549743652344,
:y2 1101.7566986083984,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1085.4890594482422,
:x2 0,
:y2 1105.572250366211,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1101.4857025146484,
:x2 0,
:y2 1121.568832397461,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 1108.1610565185547,
:x2 217.3940887451172,
:y2 1125.505599975586,
:width 1060.8,
:height 1372.8}),
:page 14},
:content {:text " This indicates that the majority of the data filtered out by CupCleaner belong to noisy data, while the data retained is predominantly in line with the intent of the comment updating task."},
:properties {:color "blue"}}
{:id #uuid "6528234b-e57e-40ec-a50c-ce2a23319b41",
:page 15,
:position {:bounding {:x1 0,
:y1 988.7318725585938,
:x2 916.3566513061523,
:y2 3153.1416015625,
:width 1060.8,
:height 1372.8},
:rects ({:x1 826.9661483764648,
:y1 988.7318725585938,
:x2 916.3566513061523,
:y2 1006.076416015625,
:width 1060.8,
:height 1372.8}
{:x1 199.53359985351562,
:y1 1012.4807739257812,
:x2 816.4725036621094,
:y2 1029.8253173828125,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 3133.058349609375,
:x2 0,
:y2 3153.1416015625,
:width 1060.8,
:height 1372.8}),
:page 15},
:content {:text "This indicates that CupCleaner can distinguish noisy data from high-quality data in comment updating datasets"},
:properties {:color "green"}}
{:id #uuid "65282382-46ef-49d3-b737-ec22905d30e2",
:page 16,
:position {:bounding {:x1 289.28331756591797,
:y1 168.34625244140625,
:x2 478.4380416870117,
:y2 185.6907958984375,
:width 1060.8,
:height 1372.8},
:rects ({:x1 289.28331756591797,
:y1 168.34625244140625,
:x2 478.4380416870117,
:y2 185.6907958984375,
:width 1060.8,
:height 1372.8}),
:page 16},
:content {:text "original, random, and cleaned"},
:properties {:color "green"}}
{:id #uuid "652823d4-f34c-4535-a478-5f27586bbc15",
:page 16,
:position {:bounding {:x1 0,
:y1 192.09515380859375,
:x2 869.9705429077148,
:y2 897.615966796875,
:width 1060.8,
:height 1372.8},
:rects ({:x1 549.8291854858398,
:y1 192.09515380859375,
:x2 869.9705429077148,
:y2 209.439697265625,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 215.8511962890625,
:x2 206.49605560302734,
:y2 233.19573974609375,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 877.5328369140625,
:x2 0,
:y2 897.615966796875,
:width 1060.8,
:height 1372.8}),
:page 16},
:content {:text "subset of the original data with the same size as the cleaned data"},
:properties {:color "green"}}
{:id #uuid "652823ea-211b-4cb8-9460-d90a7e31443a",
:page 16,
:position {:bounding {:x1 480.60790252685547,
:y1 239.60009765625,
:x2 738.5738754272461,
:y2 256.94464111328125,
:width 1060.8,
:height 1372.8},
:rects ({:x1 480.60790252685547,
:y1 239.60009765625,
:x2 738.5738754272461,
:y2 256.94464111328125,
:width 1060.8,
:height 1372.8}),
:page 16},
:content {:text " the original test set remained unchanged"},
:properties {:color "green"}}
{:id #uuid "652823f0-2a17-43c6-b49b-52874a56768e",
:page 16,
:position {:bounding {:x1 0,
:y1 239.60009765625,
:x2 870.6628036499023,
:y2 929.6092529296875,
:width 1060.8,
:height 1372.8},
:rects ({:x1 765.9308700561523,
:y1 239.60009765625,
:x2 870.6628036499023,
:y2 256.94464111328125,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 263.3489990234375,
:x2 324.2315673828125,
:y2 280.69354248046875,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 909.526123046875,
:x2 0,
:y2 929.6092529296875,
:width 1060.8,
:height 1372.8}),
:page 16},
:content {:text " we process only the training and validation sets"},
:properties {:color "yellow"}}
{:id #uuid "652824ee-8b64-401d-870b-a98d079b402a",
:page 16,
:position {:bounding {:x1 0,
:y1 693.3040161132812,
:x2 852.9213638305664,
:y2 1233.5452880859375,
:width 1060.8,
:height 1372.8},
:rects ({:x1 315.2280044555664,
:y1 693.3040161132812,
:x2 852.9213638305664,
:y2 710.6485595703125,
:width 1060.8,
:height 1372.8}
{:x1 135.88225555419922,
:y1 717.052978515625,
:x2 676.0813522338867,
:y2 734.3975219726562,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1213.462158203125,
:x2 0,
:y2 1233.5452880859375,
:width 1060.8,
:height 1372.8}),
:page 16},
:content {:text "CupCleaner shows its effectiveness regardless of whether the model performs well on the original dataset or whether the original data is heavily contaminated with noises."},
:properties {:color "green"}}
{:id #uuid "65282518-469b-4d09-8c6b-4d691d13b2b9",
:page 16,
:position {:bounding {:x1 0,
:y1 828.2520446777344,
:x2 870.0007553100586,
:y2 1297.5318603515625,
:width 1060.8,
:height 1372.8},
:rects ({:x1 537.0542526245117,
:y1 828.2520446777344,
:x2 870.0007553100586,
:y2 845.5965881347656,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 852.0009460449219,
:x2 825.2295837402344,
:y2 869.345458984375,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1277.4486694335938,
:x2 0,
:y2 1297.5318603515625,
:width 1060.8,
:height 1372.8}),
:page 16},
:content {:text " The data with scores higher than the anchor point are assigned to the cleaned version of the test set, while the rest are assigned to the noisy version of the test set."},
:properties {:color "green"}}
{:id #uuid "6528254b-bb05-469d-88b5-fedfd7543dba",
:page 17,
:position {:bounding {:x1 0,
:y1 574.5595092773438,
:x2 916.5837097167969,
:y2 1169.5587768554688,
:width 1060.8,
:height 1372.8},
:rects ({:x1 796.9305648803711,
:y1 574.5595092773438,
:x2 916.3066024780273,
:y2 591.904052734375,
:width 1060.8,
:height 1372.8}
{:x1 199.53359985351562,
:y1 598.3084106445312,
:x2 916.5837097167969,
:y2 615.6529541015625,
:width 1060.8,
:height 1372.8}
{:x1 199.53359985351562,
:y1 622.0573120117188,
:x2 616.1001281738281,
:y2 639.40185546875,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1133.4789428710938,
:x2 0,
:y2 1153.5621337890625,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1149.4755859375,
:x2 0,
:y2 1169.5587768554688,
:width 1060.8,
:height 1372.8}),
:page 17},
:content {:text "This indicates that the cleaned test set may better reflect the model’s real performance, and also shows that CupCleaner may help in constructing datasets for accurate measurement of model abilities"},
:properties {:color "green"}}
{:id #uuid "65282593-ddec-49be-a400-bc3a4fbbae40",
:page 18,
:position {:bounding {:x1 0,
:y1 287.0978698730469,
:x2 870.2934188842773,
:y2 961.6024475097656,
:width 1060.8,
:height 1372.8},
:rects ({:x1 737.4225082397461,
:y1 287.0978698730469,
:x2 870.2934188842773,
:y2 304.4424133300781,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 310.8467712402344,
:x2 604.7619323730469,
:y2 328.1913146972656,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 941.5193176269531,
:x2 0,
:y2 961.6024475097656,
:width 1060.8,
:height 1372.8}),
:page 18},
:content {:text "his indicates that our data cleaning approach can produce cleaned data within an acceptable time"},
:properties {:color "green"}}
{:id #uuid "652825e3-00c1-4b1c-81ad-ffc1424c4000",
:page 18,
:position {:bounding {:x1 0,
:y1 774.9846076965332,
:x2 870.2513961791992,
:y2 1265.5386238098145,
:width 1060.8,
:height 1372.8},
:rects ({:x1 132.09587860107422,
:y1 774.9846076965332,
:x2 870.2513961791992,
:y2 792.3291511535645,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 798.7335090637207,
:x2 306.83148193359375,
:y2 816.078052520752,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1245.4553718566895,
:x2 0,
:y2 1265.5386238098145,
:width 1060.8,
:height 1372.8}),
:page 18},
:content {:text "esults of human evaluation also demonstrate that our cleaned data aligns better with the scenarios and intentions of the comment updating tasks"},
:properties {:color "green"}}
{:id #uuid "65282606-3f49-4d98-bd46-501d37963ac4",
:page 20,
:position {:bounding {:x1 0,
:y1 273.7828063964844,
:x2 870.5523986816406,
:y2 945.6058349609375,
:width 1060.8,
:height 1372.8},
:rects ({:x1 294.00655364990234,
:y1 273.7828063964844,
:x2 869.9881820678711,
:y2 291.1273498535156,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 297.38909912109375,
:x2 870.5523986816406,
:y2 314.733642578125,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 321.13800048828125,
:x2 581.0621643066406,
:y2 338.4825439453125,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 909.5260620117188,
:x2 0,
:y2 929.6092529296875,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 925.522705078125,
:x2 0,
:y2 945.6058349609375,
:width 1060.8,
:height 1372.8}),
:page 20},
:content {:text "CupCleaner, a data cleaning approach for comment updating datasets. The data cleaning strategy of CupCleaner mainly consists of two steps. First, we design a criterion to calculate the quality scores for all data samples, and then we remove the tail of the score distribution."},
:properties {:color "purple"}}
{:id #uuid "65282611-a2df-42a5-bbea-d654edd9fda3",
:page 20,
:position {:bounding {:x1 0,
:y1 321.13800048828125,
:x2 870.2541732788086,
:y2 961.6025390625,
:width 1060.8,
:height 1372.8},
:rects ({:x1 656.8210067749023,
:y1 321.13800048828125,
:x2 870.2541732788086,
:y2 338.4825439453125,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 344.88690185546875,
:x2 850.4422302246094,
:y2 362.2314453125,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 941.5193481445312,
:x2 0,
:y2 961.6025390625,
:width 1060.8,
:height 1372.8}),
:page 20},
:content {:text "scoring process, we consider the correlations within the comments or code, as well as the correlation between code changes and comment changes"},
:properties {:color "yellow"}}
{:id #uuid "6528266e-744e-4067-a4a9-06da1fec5d40",
:page 20,
:position {:bounding {:x1 0,
:y1 439.8896484375,
:x2 870.2541885375977,
:y2 1041.585693359375,
:width 1060.8,
:height 1372.8},
:rects ({:x1 330.05374908447266,
:y1 439.8896484375,
:x2 870.2541885375977,
:y2 457.23419189453125,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 463.6385498046875,
:x2 354.2350769042969,
:y2 480.98309326171875,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1021.5025634765625,
:x2 0,
:y2 1041.585693359375,
:width 1060.8,
:height 1372.8}),
:page 20},
:content {:text "CupCleaner can effectively filter out noisy data and improve the performance of the models without changing the test set"},
:properties {:color "purple"}}
{:id #uuid "652826af-5114-42ef-9ad1-b858e9db5237",
:page 20,
:position {:bounding {:x1 0,
:y1 582.3830986022949,
:x2 870.2647933959961,
:y2 1137.5655326843262,
:width 1060.8,
:height 1372.8},
:rects ({:x1 761.4232406616211,
:y1 582.3830986022949,
:x2 870.2647933959961,
:y2 599.7276420593262,
:width 1060.8,
:height 1372.8}
{:x1 127.29556274414062,
:y1 606.1319999694824,
:x2 243.20403289794922,
:y2 623.4765434265137,
:width 1060.8,
:height 1372.8}
{:x1 0,
:y1 1117.4822807312012,
:x2 0,
:y2 1137.5655326843262,
:width 1060.8,
:height 1372.8}),
:page 20},
:content {:text "expand our data cleaning approach"},
:properties {:color "yellow"}}],
:extra {:page 19}}
Reference in New Issue View Git Blame Copy Permalink