{
  "metrics": [
    {
      "task": "t2p",
      "metric": "Sequence recovery",
      "split": "swissprot_time_cutoff",
      "model": "Chroma",
      "n": 500,
      "mean": 0.06840165630065666
    },
    {
      "task": "t2p",
      "metric": "Sequence recovery",
      "split": "swissprot_time_cutoff",
      "model": "ESM-3",
      "n": 500,
      "mean": 0.10070904955785041
    },
    {
      "task": "t2p",
      "metric": "Sequence recovery",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 0.21774230538722392
    },
    {
      "task": "t2p",
      "metric": "Sequence recovery",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-1.2B",
      "n": 500,
      "mean": 0.09249640113737709
    },
    {
      "task": "t2p",
      "metric": "Sequence recovery",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-15B",
      "n": 500,
      "mean": 0.09545050990782125
    },
    {
      "task": "t2p",
      "metric": "Sequence recovery",
      "split": "swissprot_time_cutoff",
      "model": "ProtDAT",
      "n": 500,
      "mean": 0.09740541366377221
    },
    {
      "task": "t2p",
      "metric": "Sequence recovery",
      "split": "swissprot_time_cutoff",
      "model": "ProteinDT",
      "n": 500,
      "mean": 0.07846294437013661
    },
    {
      "task": "t2p",
      "metric": "TM-score",
      "split": "swissprot_time_cutoff",
      "model": "Chroma",
      "n": 500,
      "mean": 0.28375344
    },
    {
      "task": "t2p",
      "metric": "TM-score",
      "split": "swissprot_time_cutoff",
      "model": "ESM-3",
      "n": 500,
      "mean": 0.38800424
    },
    {
      "task": "t2p",
      "metric": "TM-score",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 0.5511607199999999
    },
    {
      "task": "t2p",
      "metric": "TM-score",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-1.2B",
      "n": 500,
      "mean": 0.46487456
    },
    {
      "task": "t2p",
      "metric": "TM-score",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-15B",
      "n": 500,
      "mean": 0.49784174
    },
    {
      "task": "t2p",
      "metric": "TM-score",
      "split": "swissprot_time_cutoff",
      "model": "ProtDAT",
      "n": 500,
      "mean": 0.42719421999999996
    },
    {
      "task": "t2p",
      "metric": "TM-score",
      "split": "swissprot_time_cutoff",
      "model": "ProteinDT",
      "n": 500,
      "mean": 0.32202692
    },
    {
      "task": "t2p",
      "metric": "pLDDT",
      "split": "swissprot_time_cutoff",
      "model": "Chroma",
      "n": 500,
      "mean": 59.45439879999999
    },
    {
      "task": "t2p",
      "metric": "pLDDT",
      "split": "swissprot_time_cutoff",
      "model": "ESM-3",
      "n": 500,
      "mean": 61.755318
    },
    {
      "task": "t2p",
      "metric": "pLDDT",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 57.45689599999999
    },
    {
      "task": "t2p",
      "metric": "pLDDT",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-1.2B",
      "n": 500,
      "mean": 65.4815954
    },
    {
      "task": "t2p",
      "metric": "pLDDT",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-15B",
      "n": 500,
      "mean": 68.05155040000001
    },
    {
      "task": "t2p",
      "metric": "pLDDT",
      "split": "swissprot_time_cutoff",
      "model": "ProtDAT",
      "n": 500,
      "mean": 50.0405722
    },
    {
      "task": "t2p",
      "metric": "pLDDT",
      "split": "swissprot_time_cutoff",
      "model": "ProteinDT",
      "n": 500,
      "mean": 42.2006794
    },
    {
      "task": "p2t",
      "metric": "BLEU-4",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 54.66304
    },
    {
      "task": "p2t",
      "metric": "BLEU-4",
      "split": "swissprot_time_cutoff",
      "model": "ProtT3-ft",
      "n": 500,
      "mean": 24.51536
    },
    {
      "task": "p2t",
      "metric": "BLEU-4",
      "split": "swissprot_time_cutoff",
      "model": "ProtT3",
      "n": 500,
      "mean": 22.908839999999998
    },
    {
      "task": "p2t",
      "metric": "BLEU-4",
      "split": "swissprot_time_cutoff",
      "model": "Esm2Text-ft",
      "n": 500,
      "mean": 14.626880000000002
    },
    {
      "task": "p2t",
      "metric": "BLEU-4",
      "split": "swissprot_time_cutoff",
      "model": "Esm2Text",
      "n": 500,
      "mean": 14.649199999999999
    },
    {
      "task": "p2t",
      "metric": "BLEU-4",
      "split": "swissprot_time_cutoff",
      "model": "Galactica-ft",
      "n": 500,
      "mean": 10.981380000000001
    },
    {
      "task": "p2t",
      "metric": "BLEU-4",
      "split": "swissprot_time_cutoff",
      "model": "Galactica",
      "n": 500,
      "mean": 0.38802
    },
    {
      "task": "p2t",
      "metric": "ROUGE-L",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 65.35492572279264
    },
    {
      "task": "p2t",
      "metric": "ROUGE-L",
      "split": "swissprot_time_cutoff",
      "model": "ProtT3-ft",
      "n": 500,
      "mean": 38.67256021554208
    },
    {
      "task": "p2t",
      "metric": "ROUGE-L",
      "split": "swissprot_time_cutoff",
      "model": "ProtT3",
      "n": 500,
      "mean": 33.75807242893061
    },
    {
      "task": "p2t",
      "metric": "ROUGE-L",
      "split": "swissprot_time_cutoff",
      "model": "Esm2Text-ft",
      "n": 500,
      "mean": 30.94319533998033
    },
    {
      "task": "p2t",
      "metric": "ROUGE-L",
      "split": "swissprot_time_cutoff",
      "model": "Esm2Text",
      "n": 500,
      "mean": 31.061176609413472
    },
    {
      "task": "p2t",
      "metric": "ROUGE-L",
      "split": "swissprot_time_cutoff",
      "model": "Galactica-ft",
      "n": 500,
      "mean": 21.91380892550931
    },
    {
      "task": "p2t",
      "metric": "ROUGE-L",
      "split": "swissprot_time_cutoff",
      "model": "Galactica",
      "n": 500,
      "mean": 6.316360105563781
    },
    {
      "task": "t2p",
      "metric": "ProTrek score",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 11.416746495995671
    },
    {
      "task": "t2p",
      "metric": "ProTrek score",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-1.2B",
      "n": 500,
      "mean": 9.95387212177366
    },
    {
      "task": "t2p",
      "metric": "ProTrek score",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-15B",
      "n": 500,
      "mean": 11.283528972193599
    },
    {
      "task": "t2p",
      "metric": "ProTrek score",
      "split": "swissprot_time_cutoff",
      "model": "ESM-3",
      "n": 500,
      "mean": 3.9853699590303004
    },
    {
      "task": "t2p",
      "metric": "ProTrek score",
      "split": "swissprot_time_cutoff",
      "model": "ProteinDT",
      "n": 500,
      "mean": 4.177734241440892
    },
    {
      "task": "t2p",
      "metric": "ProTrek score",
      "split": "swissprot_time_cutoff",
      "model": "Chroma",
      "n": 500,
      "mean": 1.6105031643882395
    },
    {
      "task": "t2p",
      "metric": "ProTrek score",
      "split": "swissprot_time_cutoff",
      "model": "ProtDAT",
      "n": 500,
      "mean": 8.07194059274532
    },
    {
      "task": "t2p",
      "metric": "Keyword recovery",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 476,
      "mean": 0.59171835400827
    },
    {
      "task": "t2p",
      "metric": "Keyword recovery",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-1.2B",
      "n": 476,
      "mean": 0.4048065438296531
    },
    {
      "task": "t2p",
      "metric": "Keyword recovery",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-15B",
      "n": 476,
      "mean": 0.4911272084591412
    },
    {
      "task": "t2p",
      "metric": "Keyword recovery",
      "split": "swissprot_time_cutoff",
      "model": "ESM-3",
      "n": 476,
      "mean": 0.24700167945966264
    },
    {
      "task": "t2p",
      "metric": "Keyword recovery",
      "split": "swissprot_time_cutoff",
      "model": "ProteinDT",
      "n": 476,
      "mean": 0.10203013023391175
    },
    {
      "task": "t2p",
      "metric": "Keyword recovery",
      "split": "swissprot_time_cutoff",
      "model": "Chroma",
      "n": 476,
      "mean": 0.0
    },
    {
      "task": "t2p",
      "metric": "Keyword recovery",
      "split": "swissprot_time_cutoff",
      "model": "ProtDAT",
      "n": 476,
      "mean": 0.34750627523736766
    },
    {
      "task": "t2p",
      "metric": "Inverse folding recovery",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 0.3038814001440656
    },
    {
      "task": "t2p",
      "metric": "Inverse folding SW",
      "split": "swissprot_time_cutoff",
      "model": "Ouroboros",
      "n": 500,
      "mean": 128.686
    },
    {
      "task": "t2p",
      "metric": "Inverse folding recovery",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-1.2B",
      "n": 500,
      "mean": 0.4028382082029997
    },
    {
      "task": "t2p",
      "metric": "Inverse folding SW",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-1.2B",
      "n": 500,
      "mean": 124.632
    },
    {
      "task": "t2p",
      "metric": "Inverse folding recovery",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-15B",
      "n": 500,
      "mean": 0.4213037772012822
    },
    {
      "task": "t2p",
      "metric": "Inverse folding SW",
      "split": "swissprot_time_cutoff",
      "model": "Pinal-15B",
      "n": 500,
      "mean": 125.862
    },
    {
      "task": "t2p",
      "metric": "Inverse folding recovery",
      "split": "swissprot_time_cutoff",
      "model": "ESM-3",
      "n": 500,
      "mean": 0.3390390472018748
    },
    {
      "task": "t2p",
      "metric": "Inverse folding SW",
      "split": "swissprot_time_cutoff",
      "model": "ESM-3",
      "n": 500,
      "mean": 120.284
    },
    {
      "task": "t2p",
      "metric": "Inverse folding recovery",
      "split": "swissprot_time_cutoff",
      "model": "ProteinDT",
      "n": 500,
      "mean": 0.17933594558415994
    },
    {
      "task": "t2p",
      "metric": "Inverse folding SW",
      "split": "swissprot_time_cutoff",
      "model": "ProteinDT",
      "n": 500,
      "mean": 127.964
    },
    {
      "task": "t2p",
      "metric": "Inverse folding recovery",
      "split": "swissprot_time_cutoff",
      "model": "Chroma",
      "n": 500,
      "mean": 0.32337088516434465
    },
    {
      "task": "t2p",
      "metric": "Inverse folding SW",
      "split": "swissprot_time_cutoff",
      "model": "Chroma",
      "n": 500,
      "mean": 125.83
    },
    {
      "task": "t2p",
      "metric": "Inverse folding recovery",
      "split": "swissprot_time_cutoff",
      "model": "ProtDAT",
      "n": 500,
      "mean": 0.22630971813997303
    },
    {
      "task": "t2p",
      "metric": "Inverse folding SW",
      "split": "swissprot_time_cutoff",
      "model": "ProtDAT",
      "n": 500,
      "mean": 128.964
    }
  ],
  "scaling": [
    {
      "size": "Small",
      "parameters": 88370176,
      "recovery": 11.752343177534103,
      "n": 500
    },
    {
      "size": "Base",
      "parameters": 356310528,
      "recovery": 13.755317975732662,
      "n": 500
    },
    {
      "size": "Medium",
      "parameters": 756684288,
      "recovery": 15.291115981916583,
      "n": 500
    },
    {
      "size": "Large",
      "parameters": 1226512384,
      "recovery": 21.77423053872239,
      "n": 500
    }
  ],
  "date": "2026-10-07",
  "dataset": {
    "id": "swissprot_time_cutoff",
    "name": "Swiss-Prot time-cutoff",
    "pairs": 500,
    "keywordScoredPairs": 476,
    "sourceSplits": [
      "hard1",
      "hard"
    ]
  }
}