TabICLv2
PulseAugur coverage of TabICLv2 — every cluster mentioning TabICLv2 across labs, papers, and developer communities, ranked by signal.
6 day(s) with sentiment data
-
Tabular foundation models fail to outperform traditional methods in clinical risk prediction
A new study comparing tabular foundation models like TabPFN-3.5, TabICLv2, and Mitra-v2 against a Firth logistic regression model found that these advanced models do not offer a shortcut to accurate clinical risk predic…
-
Binary classification priors boost tabular prediction performance
A new research paper proposes a novel approach for multi-class in-context tabular prediction, demonstrating that using only binary classification priors during pretraining can achieve superior results. This method, test…
-
NVIDIA releases zero-training library for tabular data predictions
NVIDIA has launched Structured Data Models, a new PyTorch library that provides access to pretrained foundation models for tabular data. This library allows users to make predictions without requiring any additional tra…
-
New RT-J model offers context-efficient predictions on relational data
Researchers have introduced RT-J, a new 85-million-parameter relational foundation model (RFM) designed for context-efficient predictions on tabular and relational data. This model demonstrates accurate forecasting with…
-
New Relational Foundation Model RT-J Achieves Context-Efficient Predictions
Researchers have developed RT-J, an 86-million parameter relational foundation model designed for context-efficient predictions on tabular and relational data. This model requires significantly fewer examples to achieve…
-
New foundation models and benchmarks advance tabular data generation and analysis
Researchers have introduced several new foundation models and techniques for tabular data generation and analysis. TabFORGE, TabMosaic, and TabGift are new models designed for generating high-quality synthetic tabular d…
-
New research reveals spectral structures within Prior-Data Fitted Networks
Two new research papers explore the internal mechanisms of Prior-Data Fitted Networks (PFNs), a type of tabular foundation model. The first paper provides mechanistic evidence that PFNs capture spectral structures relat…
-
TIGER ensemble method sets new accuracy record for time-series classification
Researchers have developed TIGER, a novel approach to time-series classification that utilizes an ensemble of representations and adaptive meta-classification. Unlike previous state-of-the-art methods that pair bespoke …
-
New tabular foundation models TabPFN-3.5 and Causilo achieve state-of-the-art results
New tabular foundation models are advancing rapidly, with Prior Labs releasing TabPFN-3.5 and Nums AI releasing Causilo. TabPFN-3.5 demonstrates strong performance, outperforming a winning Kaggle solution from 2015 on t…
-
New attention quantization speeds up tabular foundation models
Researchers have developed a new attention quantization strategy for tabular foundation models to improve inference performance. This method focuses on quantizing queries, keys, and values to FP8, leveraging explicit FP…
-
Mitra-v2 tabular foundation model achieves SOTA performance with smaller size
Researchers have introduced Mitra-v2, a new tabular foundation model that achieves state-of-the-art performance on classification and regression tasks. Trained exclusively on synthetic data, Mitra-v2 utilizes a compact …
-
Tabular LLMs Surpass Gradient-Boosted Trees on Spreadsheet Prediction Tasks
A new class of foundation models, known as tabular LLMs, are outperforming traditional gradient-boosted trees on spreadsheet prediction tasks. These models, such as TabICLv2 and Google Research's TabFM, can predict miss…
-
Tabular foundation models fail to grasp physics principles, study finds
A new research paper investigates whether tabular foundation models (TFMs) have learned physics principles from the data they are trained on. The study evaluated four TFMs, including TabPFN-3 and TabICLv2, against six b…
-
Localized TabICLv2 improves tabular data model efficiency with k-NN retrieval
Researchers have developed Localized TabICLv2, a method to improve the efficiency of foundational models for tabular data. This new approach reduces the computational cost of TabICLv2 by only retrieving the k-nearest ne…
-
Tabular Foundation Models Show Promise but Face Deployment Hurdles
Recent research indicates that Tabular Foundation Models (TFMs), such as TabPFN and TabFM, are showing strong performance on tabular machine learning tasks, sometimes surpassing traditional gradient-boosted models like …
-
CRUMB improves PFN inference efficiency with context batching
Researchers have developed CRUMB, a novel inference wrapper designed to improve the efficiency of prior-fitted networks (PFNs). PFNs are powerful tabular foundation models that can perform in-context learning, but their…
-
Research Questions Meta-Features for Explaining Tabular Model Performance Gaps
A new research paper explores the difficulty of selecting the optimal model for tabular datasets, especially with the emergence of tabular foundation models. The study analyzed performance gaps between different model f…
-
Researchers distill large AI models into faster CPU-ready gradient-boosted trees
Researchers have developed a method to distill large tabular foundation models (TFMs) into smaller, faster gradient-boosted tree models that can run on CPUs. This technique addresses the latency issue of TFMs, which are…
-
New adapter TFM-Retouche improves tabular foundation models without fine-tuning
Researchers have developed TFM-Retouche, a novel adapter designed to enhance tabular foundation models (TFMs) without requiring computationally expensive full fine-tuning. This lightweight, architecture-agnostic adapter…