![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F16872158%2F48684c2961478bf8b3e8fbab72a77243%2F69F546CEE6A3C66CF872BA81BEB27416.png?generation=1776920915924824&alt=media\")
Hello Kagglers,
\nWe’re happy to announce the early conclusion of the Recod.ai/LUC - Scientific Image Forgery Detection Prediction competition. Thank you everyone for your participation!
\nThis competition ended with 7,268 registrations and 1,758 participants on 1,524 teams. We had 25,784 submissions from 82 countries. For 433 users (including 37 in the top 100!), this was their first competition. Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!
\nThe top potential winning teams will be contacted via email for the next steps. We look forward to learning more about their winning solutions.
\nWe cleaned the leaderboard and disqualified some teams that have violated the rules. If you think you were removed by mistake, or believe you have evidence that suggests another team cheated, please contact compliance. Please fill in all the fields honestly.
\nWe highly encourage you to post a solution write-up about your approach and solution in the forums (see instructions). You may also refer to Kaggle Solution Write-Up Documentation for guidance.
\nThank you for continuing to make Kaggle a great place to learn, practice, and test data science techniques!
", "messages": [ { "id": 3447074, "postDate": "2026-04-22T17:09:00.453Z", "content": "Hello Kagglers,
\nWe’re happy to announce the early conclusion of the Recod.ai/LUC - Scientific Image Forgery Detection Prediction competition. Thank you everyone for your participation!
\nThis competition ended with 7,268 registrations and 1,758 participants on 1,524 teams. We had 25,784 submissions from 82 countries. For 433 users (including 37 in the top 100!), this was their first competition. Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!
\nThe top potential winning teams will be contacted via email for the next steps. We look forward to learning more about their winning solutions.
\nWe cleaned the leaderboard and disqualified some teams that have violated the rules. If you think you were removed by mistake, or believe you have evidence that suggests another team cheated, please contact compliance. Please fill in all the fields honestly.
\nWe highly encourage you to post a solution write-up about your approach and solution in the forums (see instructions). You may also refer to Kaggle Solution Write-Up Documentation for guidance.
\nThank you for continuing to make Kaggle a great place to learn, practice, and test data science techniques!
", "rawMarkdown": "Hello Kagglers,\n\nWe’re happy to announce the early conclusion of the Recod.ai/LUC - Scientific Image Forgery Detection Prediction competition. Thank you everyone for your participation!\n\nThis competition ended with 7,268 registrations and 1,758 participants on 1,524 teams. We had 25,784 submissions from 82 countries. For 433 users (including 37 in the top 100!), this was their first competition. Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!\n\nThe top potential winning teams will be contacted via email for the next steps. We look forward to learning more about their winning solutions.\n\nWe cleaned the leaderboard and disqualified some teams that have violated the rules. If you think you were removed by mistake, or believe you have evidence that suggests another team cheated, please contact [compliance](https://www.kaggle.com/compliance). Please fill in all the fields honestly.\n\nWe highly encourage you to post a solution write-up about your approach and solution in the forums (see instructions). You may also refer to [Kaggle Solution Write-Up Documentation](https://www.kaggle.com/solution-write-up-documentation) for guidance.\n\nThank you for continuing to make Kaggle a great place to learn, practice, and test data science techniques!", "votes": 8 }, { "id": 3447415, "postDate": "2026-04-23T05:09:06.463Z", "content": "
I didn't see the result of the rerun, is this an official error?
", "rawMarkdown": "\n\nI didn't see the result of the rerun, is this an official error?", "votes": 1, "replies": [ { "id": 3448405, "postDate": "2026-04-25T12:13:02.570Z", "content": "I see the same. All reruns in errors. Literally all.
", "rawMarkdown": "I see the same. All reruns in errors. Literally all." }, { "id": 3475221, "postDate": "2026-06-19T16:28:13.453Z", "content": "Same here!
", "rawMarkdown": "Same here!" } ] }, { "id": 3447311, "postDate": "2026-04-22T23:36:27.590Z", "content": "@ashleyoldacre \"I have been disqualified from the competition, and my private leaderboard scores remain unrecorded. Given that I have strictly adhered to all regulations and committed no infractions whatsoever, I am seeking clarification on the precise nature of this discrepancy.\"
", "rawMarkdown": "@ashleyoldacre \"I have been disqualified from the competition, and my private leaderboard scores remain unrecorded. Given that I have strictly adhered to all regulations and committed no infractions whatsoever, I am seeking clarification on the precise nature of this discrepancy.\"", "votes": 1 }, { "id": 3449327, "postDate": "2026-04-27T22:39:04.003Z", "content": "Hi everyone,
\nI want to thank you for your patience and for the incredible effort you've put into this competition. Given the recent comments about the final leaderboard, I want to provide some clarification.
\nI know how much work goes into these models, so I’ve been in close contact with the Kaggle Team to address concerns regarding the final standings.
\nRegarding the forecasting phase, we had a success rate (solutions that ran without errors) of over 80%, which is higher than average for this type of challenge. For those whose solutions were not reflected on the leaderboard, here is some context on the likely causes:
\nMost failures occurred due to differences between the 1st phase data and the forecasting data, which caused certain pipelines to break during the rerun. As a reminder, this new data was collected from real-world forgery cases; a key part of the challenge was building solutions robust enough to handle this variability.
The Kaggle team double-checked several failure cases and confirmed there were no platform-wide issues.
We truly empathize with those whose code didn't make it through this final hurdle. Building models robust enough to handle hidden data transitions is one of the toughest parts of these competitions, especially when real-world data is so diverse.
\nWe sincerely appreciate the dedication you showed throughout the process. We hope this doesn't discourage you, as this field needs strong minds like yours.
\nBefore signing off, I also want to express my gratitude to the Kaggle team, who have worked incredibly hard to make this competition run as smoothly as possible.
\nThank you for being part of this community!
\nKind regards,
\nJoão Phillipe
", "rawMarkdown": "Hi everyone,\n\nI want to thank you for your patience and for the incredible effort you've put into this competition. Given the recent comments about the final leaderboard, I want to provide some clarification.\n\nI know how much work goes into these models, so I’ve been in close contact with the Kaggle Team to address concerns regarding the final standings.\n\nRegarding the forecasting phase, we had a success rate (solutions that ran without errors) of over 80%, which is higher than average for this type of challenge. For those whose solutions were not reflected on the leaderboard, here is some context on the likely causes:\n\n- Most failures occurred due to differences between the 1st phase data and the forecasting data, which caused certain pipelines to break during the rerun. As a reminder, this new data was collected from real-world forgery cases; a key part of the challenge was building solutions robust enough to handle this variability.\n\n- The Kaggle team double-checked several failure cases and confirmed there were no platform-wide issues.\n\nWe truly empathize with those whose code didn't make it through this final hurdle. Building models robust enough to handle hidden data transitions is one of the toughest parts of these competitions, especially when real-world data is so diverse.\n\nWe sincerely appreciate the dedication you showed throughout the process. We hope this doesn't discourage you, as this field needs strong minds like yours.\n\nBefore signing off, I also want to express my gratitude to the **Kaggle team**, who have worked incredibly hard to make this competition run as smoothly as possible.\n\nThank you for being part of this community!\n\nKind regards,\n\nJoão Phillipe", "votes": 2, "replies": [ { "id": 3449329, "postDate": "2026-04-27T22:49:05.770Z", "content": "This is surprising. Thanks for the advice.😂😂
", "rawMarkdown": "This is surprising. Thanks for the advice.😂😂" } ] }, { "id": 3447347, "postDate": "2026-04-23T01:05:36.103Z", "content": "@ashleyoldacre @joophillipecardenuto \nI was also removed from the private leaderboard.
\nI have no awareness of any rule violation, so I suspect my submission may have failed due to an error. If a submission fails because of an error, is it also removed from the leaderboard?
", "rawMarkdown": "@ashleyoldacre @joophillipecardenuto \nI was also removed from the private leaderboard.\n\nI have no awareness of any rule violation, so I suspect my submission may have failed due to an error. If a submission fails because of an error, is it also removed from the leaderboard?", "votes": 2 }, { "id": 3448489, "postDate": "2026-04-25T16:54:05.467Z", "content": "Yeah my scores are also not.
", "rawMarkdown": "Yeah my scores are also not." }, { "id": 3447756, "postDate": "2026-04-23T20:53:34.390Z", "content": "@ashleyoldacre @joophillipecardenuto same here, I'm not showing up on the leaderboard at all. Could you help me understand why that might be?
", "rawMarkdown": "@ashleyoldacre @joophillipecardenuto same here, I'm not showing up on the leaderboard at all. Could you help me understand why that might be?" }, { "id": 3447115, "postDate": "2026-04-22T18:12:18.017Z", "content": "Is it complete?
", "rawMarkdown": "Is it complete?\n", "replies": [ { "id": 3447260, "postDate": "2026-04-22T21:37:20.617Z", "content": "Yes, the leaderboard is complete and finalize.
", "rawMarkdown": "Yes, the leaderboard is complete and finalize. " } ] }, { "id": 3447338, "postDate": "2026-04-23T00:37:50.570Z", "rawMarkdown": "", "isDeleted": true, "replies": [ { "id": 3447341, "postDate": "2026-04-23T00:50:12.030Z", "rawMarkdown": "", "isDeleted": true } ] }, { "id": 3447079, "postDate": "2026-04-22T17:15:15.840Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 3447415, "author_name": "WentJiao", "author_url": "", "post_date": "2026-04-23T05:09:06.463000", "content": "
I didn't see the result of the rerun, is this an official error?
", "votes": 1, "replies": [ { "id": 3448405, "author_name": "Ivan Ilyushchenko", "author_url": "", "post_date": "2026-04-25T12:13:02.570000", "content": "I see the same. All reruns in errors. Literally all.
", "votes": 0, "replies": [] }, { "id": 3475221, "author_name": "Davide Fiocco", "author_url": "", "post_date": "2026-06-19T16:28:13.453000", "content": "Same here!
", "votes": 0, "replies": [] } ] }, { "id": 3447311, "author_name": "THE KING 👑", "author_url": "", "post_date": "2026-04-22T23:36:27.590000", "content": "@ashleyoldacre \"I have been disqualified from the competition, and my private leaderboard scores remain unrecorded. Given that I have strictly adhered to all regulations and committed no infractions whatsoever, I am seeking clarification on the precise nature of this discrepancy.\"
", "votes": 1, "replies": [] }, { "id": 3449327, "author_name": "João Phillipe Cardenuto", "author_url": "", "post_date": "2026-04-27T22:39:04.003000", "content": "Hi everyone,
\nI want to thank you for your patience and for the incredible effort you've put into this competition. Given the recent comments about the final leaderboard, I want to provide some clarification.
\nI know how much work goes into these models, so I’ve been in close contact with the Kaggle Team to address concerns regarding the final standings.
\nRegarding the forecasting phase, we had a success rate (solutions that ran without errors) of over 80%, which is higher than average for this type of challenge. For those whose solutions were not reflected on the leaderboard, here is some context on the likely causes:
\nMost failures occurred due to differences between the 1st phase data and the forecasting data, which caused certain pipelines to break during the rerun. As a reminder, this new data was collected from real-world forgery cases; a key part of the challenge was building solutions robust enough to handle this variability.
The Kaggle team double-checked several failure cases and confirmed there were no platform-wide issues.
We truly empathize with those whose code didn't make it through this final hurdle. Building models robust enough to handle hidden data transitions is one of the toughest parts of these competitions, especially when real-world data is so diverse.
\nWe sincerely appreciate the dedication you showed throughout the process. We hope this doesn't discourage you, as this field needs strong minds like yours.
\nBefore signing off, I also want to express my gratitude to the Kaggle team, who have worked incredibly hard to make this competition run as smoothly as possible.
\nThank you for being part of this community!
\nKind regards,
\nJoão Phillipe
", "votes": 2, "replies": [ { "id": 3449329, "author_name": "THE KING 👑", "author_url": "", "post_date": "2026-04-27T22:49:05.770000", "content": "This is surprising. Thanks for the advice.😂😂
", "votes": 0, "replies": [] } ] }, { "id": 3447347, "author_name": "moritake04", "author_url": "", "post_date": "2026-04-23T01:05:36.103000", "content": "@ashleyoldacre @joophillipecardenuto \nI was also removed from the private leaderboard.
\nI have no awareness of any rule violation, so I suspect my submission may have failed due to an error. If a submission fails because of an error, is it also removed from the leaderboard?
", "votes": 2, "replies": [] }, { "id": 3448489, "author_name": "GAURAV PARKHEDKAR", "author_url": "", "post_date": "2026-04-25T16:54:05.467000", "content": "Yeah my scores are also not.
", "votes": 0, "replies": [] }, { "id": 3447756, "author_name": "Azat Akhtyamov", "author_url": "", "post_date": "2026-04-23T20:53:34.390000", "content": "@ashleyoldacre @joophillipecardenuto same here, I'm not showing up on the leaderboard at all. Could you help me understand why that might be?
", "votes": 0, "replies": [] }, { "id": 3447115, "author_name": "Baidalin Adilzhan [dsmlkz]", "author_url": "", "post_date": "2026-04-22T18:12:18.017000", "content": "Is it complete?
", "votes": 0, "replies": [ { "id": 3447260, "author_name": "Ashley Oldacre", "author_url": "", "post_date": "2026-04-22T21:37:20.617000", "content": "Yes, the leaderboard is complete and finalize.
", "votes": 0, "replies": [] } ] }, { "id": 3447338, "author_name": "", "author_url": "", "post_date": "2026-04-23T00:37:50.570000", "content": "", "votes": 0, "replies": [ { "id": 3447341, "author_name": "", "author_url": "", "post_date": "2026-04-23T00:50:12.030000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 3447079, "author_name": "", "author_url": "", "post_date": "2026-04-22T17:15:15.840000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "3447074": "Hello Kagglers,\n\nWe’re happy to announce the early conclusion of the Recod.ai/LUC - Scientific Image Forgery Detection Prediction competition. Thank you everyone for your participation!\n\nThis competition ended with 7,268 registrations and 1,758 participants on 1,524 teams. We had 25,784 submissions from 82 countries. For 433 users (including 37 in the top 100!), this was their first competition. Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!\n\nThe top potential winning teams will be contacted via email for the next steps. We look forward to learning more about their winning solutions.\n\nWe cleaned the leaderboard and disqualified some teams that have violated the rules. If you think you were removed by mistake, or believe you have evidence that suggests another team cheated, please contact [compliance](https://www.kaggle.com/compliance). Please fill in all the fields honestly.\n\nWe highly encourage you to post a solution write-up about your approach and solution in the forums (see instructions). You may also refer to [Kaggle Solution Write-Up Documentation](https://www.kaggle.com/solution-write-up-documentation) for guidance.\n\nThank you for continuing to make Kaggle a great place to learn, practice, and test data science techniques!", "3447415": "\n\nI didn't see the result of the rerun, is this an official error?", "3447311": "@ashleyoldacre \"I have been disqualified from the competition, and my private leaderboard scores remain unrecorded. Given that I have strictly adhered to all regulations and committed no infractions whatsoever, I am seeking clarification on the precise nature of this discrepancy.\"", "3449327": "Hi everyone,\n\nI want to thank you for your patience and for the incredible effort you've put into this competition. Given the recent comments about the final leaderboard, I want to provide some clarification.\n\nI know how much work goes into these models, so I’ve been in close contact with the Kaggle Team to address concerns regarding the final standings.\n\nRegarding the forecasting phase, we had a success rate (solutions that ran without errors) of over 80%, which is higher than average for this type of challenge. For those whose solutions were not reflected on the leaderboard, here is some context on the likely causes:\n\n- Most failures occurred due to differences between the 1st phase data and the forecasting data, which caused certain pipelines to break during the rerun. As a reminder, this new data was collected from real-world forgery cases; a key part of the challenge was building solutions robust enough to handle this variability.\n\n- The Kaggle team double-checked several failure cases and confirmed there were no platform-wide issues.\n\nWe truly empathize with those whose code didn't make it through this final hurdle. Building models robust enough to handle hidden data transitions is one of the toughest parts of these competitions, especially when real-world data is so diverse.\n\nWe sincerely appreciate the dedication you showed throughout the process. We hope this doesn't discourage you, as this field needs strong minds like yours.\n\nBefore signing off, I also want to express my gratitude to the **Kaggle team**, who have worked incredibly hard to make this competition run as smoothly as possible.\n\nThank you for being part of this community!\n\nKind regards,\n\nJoão Phillipe", "3447347": "@ashleyoldacre @joophillipecardenuto \nI was also removed from the private leaderboard.\n\nI have no awareness of any rule violation, so I suspect my submission may have failed due to an error. If a submission fails because of an error, is it also removed from the leaderboard?", "3448489": "Yeah my scores are also not.", "3447756": "@ashleyoldacre @joophillipecardenuto same here, I'm not showing up on the leaderboard at all. Could you help me understand why that might be?", "3447115": "Is it complete?\n", "3447338": "", "3447079": "" } }