{
  "annotation_id": "2c09a842-40dc-4aa4-b9c0-1d0a5f12d974",
  "date_created": "2026-02-17T05:53:20.447000Z",
  "date_modified": "2026-02-17T05:53:20.447000Z",
  "file_hash": "10ce518098e225a5672edc57c9462076e5c977da0143c2569b36bed344db65e2",
  "private": false,
  "record": {
    "abstract": "Tabular Foundation Models (TFMs) have recently shown strong in-context learning capabilities on structured data, achieving zero-shot performance comparable to traditional machine learning methods. We find that zero-shot TFMs already achieve strong performance, while the benefits of fine-tuning are highly model and data-dependent. Meta-learning and PEFT provide moderate gains under specific conditions, whereas full supervised fine-tuning (SFT) often reduces accuracy or calibration quality. This work presents the first comprehensive study of fine-tuning in TFMs across benchmarks including TALENT, OpenML-CC18, and TabZilla. We compare Zero-Shot, Meta-Learning, Supervised (SFT), and parameter-efficient (PEFT) approaches, analyzing how dataset factors such as imbalance, size, and dimensionality affect outcomes. Our findings cover performance, calibration, and fairness, offering practical guidelines on when fine-tuning is most beneficial and its limitations.",
    "arxiv_id": "2601.09654",
    "authors": [
      "Aditya Tanna",
      "Pratinav Seth",
      "Mohamed Bouadi",
      "Vinay Kumar Sankarapu"
    ],
    "categories": [
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "title": "Exploring Fine-Tuning for Tabular Foundation Models",
    "url": "https://arxiv.org/abs/2601.09654",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "6eb6aafb-d4f5-4520-8061-a7378c0013c4",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}