<a href=\"https://cdn-uploads.huggingface.co/production/uploads/67039e7443929668458d3618/G3-DkG1OiDnT0qHQkNWrE.png\" rel=\"nofollow\"><img src=\"https://cdn-uploads.huggingface.co/production/uploads/67039e7443929668458d3618/G3-DkG1OiDnT0qHQkNWrE.png\" alt=\"main_figure\"></a></p>\n<p>This paper proposes Adaptive Binning for medical tabular self-supervised learning. The core idea is to replace fixed global quantile binning with a learning-coupled, feature-wise coarse-to-fine curriculum that determines when to refine each feature, where to split its bins, and how to supervise mixed categorical–numerical schemas through type-aware ordinal reconstruction.</p>\n<p>We show that adaptive discretization yields stronger representations across diverse public medical tabular datasets in both linear probing and fine-tuning evaluations. We also establish a unified benchmark for reproducible medical tabular self-supervised learning.</p>\n","updatedAt":"2026-06-22T21:45:03.225Z","author":{"_id":"67039e7443929668458d3618","avatarUrl":"/avatars/081486010c5d27153410f68cc3325e6a.svg","fullname":"DaehwanKim","name":"HYUDHKIM","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8518046140670776},"editors":["HYUDHKIM"],"editorAvatarUrls":["/avatars/081486010c5d27153410f68cc3325e6a.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.19827","authors":[{"_id":"6a3891efdb23715e9da138df","user":{"_id":"67039e7443929668458d3618","avatarUrl":"/avatars/081486010c5d27153410f68cc3325e6a.svg","isPro":false,"fullname":"DaehwanKim","user":"HYUDHKIM","type":"user","name":"HYUDHKIM"},"name":"Daehwan Kim","status":"claimed_verified","statusLastChangedAt":"2026-06-22T16:12:34.063Z","hidden":false},{"_id":"6a3891efdb23715e9da138e0","name":"Haejun Chung","hidden":false},{"_id":"6a3891efdb23715e9da138e1","name":"Ikbeom Jang","hidden":false}],"publishedAt":"2026-06-18T00:00:00.000Z","submittedOnDailyAt":"2026-06-22T00:00:00.000Z","title":"When, Where, and How: Adaptive Binning for Tabular Self-Supervised Learning","submittedOnDailyBy":{"_id":"67039e7443929668458d3618","avatarUrl":"/avatars/081486010c5d27153410f68cc3325e6a.svg","isPro":false,"fullname":"DaehwanKim","user":"HYUDHKIM","type":"user","name":"HYUDHKIM"},"summary":"Medical tabular data are ubiquitous in clinical research, but deep learning for tables remains underexplored because reliable labels often require costly expert adjudication, even though structured clinical variables are routinely available in tabular form. Self-supervised learning can leverage these unlabeled tables, and recent binning-based pretexts offer a promising inductive bias, but existing objectives fix a single global quantile discretization and apply feature-agnostic supervision. We propose Adaptive Binning, a training-adaptive discretization pretext for tabular SSL that couples discretization to learning through a feature-wise coarse-to-fine curriculum. Motivated by the spectral bias of neural networks and the principles of curriculum learning, our method progressively refines discretization per feature upon plateau detection and selects representation-aware splits to jointly improve value-space concentration and representation-space coherence. A heterogeneity-aware objective unifies categorical reconstruction with ordinal supervision for numerical features, and experiments on public medical tabular datasets under unified evaluation protocols show consistent gains for linear probing and fine-tuning without dataset-specific discretization tuning. We further introduce a medical tabular SSL benchmark with standardized protocols to support reproducible progress in this underexplored domain. Our code is available at https://github.com/labhai/Adaptive-Binning.","upvotes":2,"discussionId":"6a3891efdb23715e9da138e2","projectPage":"https://github.com/labhai/Adaptive-Binning","githubRepo":"https://github.com/labhai/Adaptive-Binning","githubRepoAddedBy":"user","ai_summary":"Adaptive Binning introduces a training-adaptive discretization method for self-supervised learning on medical tabular data, improving representation learning through feature-wise refinement and heterogeneous feature handling.","ai_keywords":["self-supervised learning","tabular data","discretization","pretexts","spectral bias","curriculum learning","fine-tuning","linear probing","medical tabular SSL","heterogeneous-aware objective","feature-wise coarse-to-fine curriculum"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":0},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"67039e7443929668458d3618","avatarUrl":"/avatars/081486010c5d27153410f68cc3325e6a.svg","isPro":false,"fullname":"DaehwanKim","user":"HYUDHKIM","type":"user"},{"_id":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.19827.md","query":{}}">
When, Where, and How: Adaptive Binning for Tabular Self-Supervised Learning
Abstract
Adaptive Binning introduces a training-adaptive discretization method for self-supervised learning on medical tabular data, improving representation learning through feature-wise refinement and heterogeneous feature handling.
Medical tabular data are ubiquitous in clinical research, but deep learning for tables remains underexplored because reliable labels often require costly expert adjudication, even though structured clinical variables are routinely available in tabular form. Self-supervised learning can leverage these unlabeled tables, and recent binning-based pretexts offer a promising inductive bias, but existing objectives fix a single global quantile discretization and apply feature-agnostic supervision. We propose Adaptive Binning, a training-adaptive discretization pretext for tabular SSL that couples discretization to learning through a feature-wise coarse-to-fine curriculum. Motivated by the spectral bias of neural networks and the principles of curriculum learning, our method progressively refines discretization per feature upon plateau detection and selects representation-aware splits to jointly improve value-space concentration and representation-space coherence. A heterogeneity-aware objective unifies categorical reconstruction with ordinal supervision for numerical features, and experiments on public medical tabular datasets under unified evaluation protocols show consistent gains for linear probing and fine-tuning without dataset-specific discretization tuning. We further introduce a medical tabular SSL benchmark with standardized protocols to support reproducible progress in this underexplored domain. Our code is available at https://github.com/labhai/Adaptive-Binning.
Community
This paper proposes Adaptive Binning for medical tabular self-supervised learning. The core idea is to replace fixed global quantile binning with a learning-coupled, feature-wise coarse-to-fine curriculum that determines when to refine each feature, where to split its bins, and how to supervise mixed categorical–numerical schemas through type-aware ordinal reconstruction.
We show that adaptive discretization yields stronger representations across diverse public medical tabular datasets in both linear probing and fine-tuning evaluations. We also establish a unified benchmark for reproducible medical tabular self-supervised learning.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2606.19827 in a model README.md to link it from this page.
Cite arxiv.org/abs/2606.19827 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.19827 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.