Topic
tabular data
DataMagic System Transforms Raw Tabular Data into Narrative Data-Insight Videos
DataMagic, an interactive system, transforms raw tabular data and natural language queries into narrative data-insight videos. It uses DVSpec to bind visuals to data fields and a Generate-then-Orchestrate architecture to optimize narrative coherence. Evaluated on 109 real-world samples, the system enables explorable interactive data interfaces.
Adaptive Binning Boosts Self-Supervised Learning on Medical Tabular Data, Researchers Report
Researchers propose Adaptive Binning, a training-adaptive discretization pretext for tabular self-supervised learning. The method progressively refines discretization per feature and uses a heterogeneity-aware objective. Experiments on public medical tabular datasets show consistent gains over fixed binning approaches.
LLM Confidence Is Epistemically Vacuous: New Method Detects Blind Spots in Clinical Data
A new study reveals that large language models (LLMs) fail to recognize their own knowledge limits on structured clinical data, outputting near-constant confidence scores regardless of accuracy. Researchers propose a cross-model calibrator using attribution divergence between LLM and XGBoost, reducing calibration error from 0.254 to 0.080 and improving accuracy from 49% to 75.3% without training.
LLMs Struggle on Privacy-Constrained Industrial Tabular Data, Study Finds
A new study from arXiv compares large language models (LLMs) with classical machine learning on an industrial car retrofit prediction task, finding that while LLMs have niche uses, tree ensembles remain superior. The research highlights that on privacy-constrained tables, LLMs are more effective as complementary components than replacements.