{
  "created_utc": "2026-09-23T09:07:49.626404+00:00",
  "sample_count": 100,
  "group_count": 37,
  "folds": 5,
  "seeds": [
    42,
    3407,
    2026
  ],
  "device": "cuda",
  "gpu_name": "NVIDIA GeForce RTX 5070 Ti",
  "python": "3.14.7",
  "torch": "2.14.0+cu130",
  "parameters": {
    "grid_size": 50,
    "hidden_size": 128,
    "heads": 4,
    "dropout": 0.1,
    "batch_size": 8,
    "epochs_max": 50,
    "early_stopping_patience": 8,
    "learning_rate": 0.0001,
    "mask_ratio": 0.2,
    "retrieval_probe_epochs": 20,
    "reconstruction_probe_epochs": 25,
    "mvr_epsilon": 0.02
  },
  "objective": "masked reconstruction + cross-modal contrastive + temporal monotonicity; emotion labels unused",
  "split_rule": "GroupKFold by group_id/video_id",
  "example_sample_id": "-tPCytz4rww/12",
  "elapsed_seconds": 516.2944533824921,
  "interpretation_limits": [
    "Grid-index retrieval is a representation-consistency probe, not independent temporal ground truth.",
    "Masked reconstruction uses a decoder trained on the training fold and reports standardized-feature errors.",
    "The emotion probe is a small-sample downstream utility check, not a claim of generalization to MOSEI.",
    "No human event timestamps are available, so human IoU/MATE is not reported."
  ]
}