{
  "id": 696664,
  "title": "Phase Two Finalists of the Dishcovery Mission II [CVPR 2026]",
  "url": "/competitions/dishcovery-mission-ii-cvpr-2026/discussion/696664",
  "author_name": "",
  "post_date": "2026-05-03T10:43:09.273744600Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>The first phase of Dishcovery Mission II [CVPR 2026] has concluded! <strong>Thank you to all participants</strong> for your submissions and hard work throughout the challenge.</p>\n<p>We are pleased to announce that the following four teams will advance to Phase Two:</p>\n<ol>\n<li>trytry4625 (score: 0.78966)</li>\n<li>Atromx (score: 0.71984)</li>\n<li>klyg_0_0 (score: 0.69341)</li>\n<li>fightraccoon (score: 0.69270)</li>\n</ol>\n<p>As described in the challenge rules, Phase Two will focus on <strong>reproducibility and final evaluation</strong>. We will reproduce the Phase One results from the submitted model and scripts. Non-reproducible results may lead to disqualification. Then, <strong>all selected teams will be evaluated on a private test set to determine the final winner and runner-up</strong>.</p>\n<p>Each finalist team must send us the following materials:</p>\n<ol>\n<li>The <strong>complete codebase</strong> used to obtain the leaderboard result.</li>\n<li>The <strong>exact model checkpoint/weights</strong> corresponding to the submitted result.</li>\n<li><strong>Detailed instructions explaining how to run the code</strong>, including all dependencies and versions.</li>\n<li><strong>Environment setup</strong> instructions.</li>\n<li><strong>Preprocessing and inference</strong> commands if needed.</li>\n<li>Any <strong>quantization or special configuration required</strong> for evaluation.</li>\n<li>A .txt file listing all <strong>sources of information/data</strong> used to train the model.</li>\n</ol>\n<p>Please send all required materials by Wednesday, May 6, to <strong><em>mtf25.vlm.challenge@gmail.com</em></strong></p>\n<p>Congratulations to the four advancing teams!!</p>",
  "messages": [
    {
      "id": "3452428",
      "postDate": "05/03/2026 10:43:09",
      "content": "<p>The first phase of Dishcovery Mission II [CVPR 2026] has concluded! <strong>Thank you to all participants</strong> for your submissions and hard work throughout the challenge.</p>\n<p>We are pleased to announce that the following four teams will advance to Phase Two:</p>\n<ol>\n<li>trytry4625 (score: 0.78966)</li>\n<li>Atromx (score: 0.71984)</li>\n<li>klyg_0_0 (score: 0.69341)</li>\n<li>fightraccoon (score: 0.69270)</li>\n</ol>\n<p>As described in the challenge rules, Phase Two will focus on <strong>reproducibility and final evaluation</strong>. We will reproduce the Phase One results from the submitted model and scripts. Non-reproducible results may lead to disqualification. Then, <strong>all selected teams will be evaluated on a private test set to determine the final winner and runner-up</strong>.</p>\n<p>Each finalist team must send us the following materials:</p>\n<ol>\n<li>The <strong>complete codebase</strong> used to obtain the leaderboard result.</li>\n<li>The <strong>exact model checkpoint/weights</strong> corresponding to the submitted result.</li>\n<li><strong>Detailed instructions explaining how to run the code</strong>, including all dependencies and versions.</li>\n<li><strong>Environment setup</strong> instructions.</li>\n<li><strong>Preprocessing and inference</strong> commands if needed.</li>\n<li>Any <strong>quantization or special configuration required</strong> for evaluation.</li>\n<li>A .txt file listing all <strong>sources of information/data</strong> used to train the model.</li>\n</ol>\n<p>Please send all required materials by Wednesday, May 6, to <strong><em>mtf25.vlm.challenge@gmail.com</em></strong></p>\n<p>Congratulations to the four advancing teams!!</p>",
      "rawMarkdown": "The first phase of Dishcovery Mission II [CVPR 2026] has concluded! **Thank you to all participants** for your submissions and hard work throughout the challenge.\n\nWe are pleased to announce that the following four teams will advance to Phase Two:\n\n1.\ttrytry4625 (score: 0.78966)\n2.\tAtromx (score: 0.71984)\n3.\tklyg_0_0 (score: 0.69341)\n4.\tfightraccoon (score: 0.69270)\n\nAs described in the challenge rules, Phase Two will focus on **reproducibility and final evaluation**. We will reproduce the Phase One results from the submitted model and scripts. Non-reproducible results may lead to disqualification. Then, **all selected teams will be evaluated on a private test set to determine the final winner and runner-up**.\n\nEach finalist team must send us the following materials:\n\n1.\tThe **complete codebase** used to obtain the leaderboard result.\n2.\tThe **exact model checkpoint/weights** corresponding to the submitted result.\n3.\t**Detailed instructions explaining how to run the code**, including all dependencies and versions.\n4.\t**Environment setup** instructions.\n5.\t**Preprocessing and inference** commands if needed.\n6.\tAny **quantization or special configuration required** for evaluation.\n7.\tA .txt file listing all **sources of information/data** used to train the model.\n\nPlease send all required materials by Wednesday, May 6, to ***mtf25.vlm.challenge@gmail.com***\n\nCongratulations to the four advancing teams!!",
      "votes": null
    },
    {
      "id": "3453898",
      "postDate": "05/06/2026 06:11:26",
      "content": "<p>When will we know the  final winner?</p>",
      "rawMarkdown": "When will we know the  final winner?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3453898,
      "author_name": "mikechen4625",
      "author_url": "",
      "post_date": "05/06/2026 06:11:26",
      "content": "<p>When will we know the  final winner?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3452428": "The first phase of Dishcovery Mission II [CVPR 2026] has concluded! **Thank you to all participants** for your submissions and hard work throughout the challenge.\n\nWe are pleased to announce that the following four teams will advance to Phase Two:\n\n1.\ttrytry4625 (score: 0.78966)\n2.\tAtromx (score: 0.71984)\n3.\tklyg_0_0 (score: 0.69341)\n4.\tfightraccoon (score: 0.69270)\n\nAs described in the challenge rules, Phase Two will focus on **reproducibility and final evaluation**. We will reproduce the Phase One results from the submitted model and scripts. Non-reproducible results may lead to disqualification. Then, **all selected teams will be evaluated on a private test set to determine the final winner and runner-up**.\n\nEach finalist team must send us the following materials:\n\n1.\tThe **complete codebase** used to obtain the leaderboard result.\n2.\tThe **exact model checkpoint/weights** corresponding to the submitted result.\n3.\t**Detailed instructions explaining how to run the code**, including all dependencies and versions.\n4.\t**Environment setup** instructions.\n5.\t**Preprocessing and inference** commands if needed.\n6.\tAny **quantization or special configuration required** for evaluation.\n7.\tA .txt file listing all **sources of information/data** used to train the model.\n\nPlease send all required materials by Wednesday, May 6, to ***mtf25.vlm.challenge@gmail.com***\n\nCongratulations to the four advancing teams!!",
    "3453898": "When will we know the  final winner?"
  },
  "source": "meta"
}