tabular foundation models
PulseAugur coverage of tabular foundation models — every cluster mentioning tabular foundation models across labs, papers, and developer communities, ranked by signal.
- 2026-05-18 research_milestone A new paper details a method for distilling tabular foundation models for structured health data. source
5 day(s) with sentiment data
-
Google Research's TabFM outperforms LLMs in financial forecasting
Researchers have introduced tabular foundation models (TFMs) as a promising new tool for financial forecasting, particularly for predicting firms' next-year earnings changes. A TFM developed by Google Research, named Ta…
-
New research explores efficient data labeling for tabular foundation models
A new research paper introduces active in-context learning techniques for tabular foundation models (TFMs). The study, titled "When, Where, How: Active In-Context Learning for Tabular Foundation Models," explores how to…
-
MulTaBench benchmark advances multimodal tabular learning with text and image
Researchers have introduced MulTaBench, a new benchmark designed to evaluate multimodal tabular learning, which combines text and image data with structured tables. The benchmark consists of 40 datasets focused on tasks…
-
State Estimation framed as In-context Tabular Prediction using Foundation Models
Researchers have proposed a novel approach to state estimation by framing it as an in-context tabular prediction problem. This method utilizes pre-trained tabular foundation models with frozen weights, requiring only hi…
-
New MICE method tackles concept drift in tabular foundation models
Researchers have developed MICE (Mixture of In-Context Experts), a novel training-free method for tabular foundation models (TFMs) that operate on data streams. MICE addresses the challenge of recurring concept drift by…
-
LLM agent RelAgent acts as autonomous data scientist for relational learning
Researchers have introduced RelAgent, an LLM-based autonomous data scientist designed for relational learning tasks. This system operates in two phases: a search phase where an LLM agent constructs SQL feature programs …
-
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…
-
Research finds Tabular Foundation Models are overparameterized, suggesting simplification
A new research paper suggests that Tabular Foundation Models (TFMs), which use transformer architectures, are significantly overparameterized. The study introduces a method to replace up to 96% of a TFM's blocks without…
-
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…
-
New method improves training-free node classification with tabular foundation models
Researchers have introduced HG-DAPS, a novel training-free diffusion score designed for node classification in tabular foundation models. This method utilizes a homophily gate that processes only in-context labels, ensu…
-
New TICDA method attributes data influence in tabular foundation models
Researchers have developed TICDA, a novel method for attributing the influence of individual data points within the context of tabular foundation models (TFMs). Unlike existing methods that are computationally expensive…
-
New certificates reveal joint processing in tabular foundation models
Researchers have identified a "standardization trap" in the evaluation of pretrained tabular foundation models (TFMs). This trap arises because public TFM packages standardize labels before the model processes them, obs…
-
Activation alignment boosts tabular model in-context learning
Researchers have developed a new method called activation alignment to improve the performance of tabular foundation models during in-context learning. This technique trains a lightweight linear transformation to map th…
-
New research advances tabular foundation models for efficiency and robustness
Multiple research papers explore advancements in tabular foundation models (TFMs), focusing on improving their efficiency, robustness, and adaptability. Techniques such as knowledge distillation are being used to create…
-
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…
-
New HINT framework optimizes Tabular Foundation Models for data streams
Researchers have introduced HINT, a novel framework designed to improve the efficiency of Tabular Foundation Models (TFMs) in high-throughput data streams. HINT addresses challenges related to communication overhead and…
-
Research paper analyzes synthetic data effectiveness for tabular foundation models
A new research paper explores the effectiveness of synthetic data used in pretraining tabular foundation models. The study analyzes how well these synthetic data generators support downstream tasks by comparing their ge…
-
New interfaces enhance tabular foundation models for time-to-event prediction
Researchers have developed new adaptation interfaces to improve the performance of tabular foundation models (TabFMs) in time-to-event prediction tasks. These interfaces address the challenges of handling censored data …
-
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…
-
ARASH method boosts TFM efficiency for tabular prediction
Researchers have developed ARASH, a novel method designed to improve the efficiency of tabular foundation models (TFMs) like TabPFN. ARASH addresses the challenge of selecting optimal few-shot examples for tabular data …