2601.20448v1.pdf

{
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "0d91c2e3629ee9af9ed414f834e3af5a02246d45d8e72a6494c2fa84826da814",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.20448v1.pdf",
      "size": 3922536
    },
    "source": {
      "execution_id": "8c177e36-4785-4db4-b817-07896c4d2d8d",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "creation_date": "2026-01-29T01:49:01Z",
      "keywords": [
        "Variational Autoencoder",
        "Time Series Forecasting",
        "Sudden Fluctuations Enhancement",
        "Latent Representation Learning"
      ],
      "modified_date": "2026-01-29T01:49:01Z",
      "page_count": 13,
      "producer": "pikepdf 8.15.1",
      "subject": "-  Computing methodologies  -\u003e  Machine learning.",
      "title": "TimeCatcher: A Variational Framework for Volatility-Aware Forecasting of Non-Stationary Time Series",
      "version": "1.7"
    },
    "source": {
      "execution_id": "b0d9cec8-17c3-4de2-a962-7721e06a8038",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}