TabICLv2 is a new state‑of‑the‑art tabular foundation model that outperforms existing methods on regression and classification tasks. It relies on a synthetic data generation engine for diverse pretraining, architectural innovations such as a scalable softmax attention, and optimized training protocols that replace AdamW with the Muon optimizer. On the TabArena and TALENT benchmarks, TabICLv2 surpasses the current best model, RealTabPFN‑2.5, without any tuning, while also being faster and capable of handling million‑scale datasets with limited GPU memory.
By Jingang Qu, David Holzm\"uller, Ga\"el Varoquaux, Marine Le Morvan
arXiv:2607. 27546v1 Announce Type: new Abstract: Tabular Foundation Models, such as TabPFN, have received a large amount of recent attention due to their performance on in-context tabular machine learning tasks, which often exceeds classical baselines.
By Shuting Luo, Monika Mikhail Kanaan, Cameron Gordon, Anna Leontjeva, Simon Lucey
arXiv:2607. 05380v1 Announce Type: new Abstract: In deep learning for tabular data, efficient ensembles of multilayer perceptrons (MLPs) have recently emerged as effective and practical architectures.
By Yury Gorishniy, Akim Kotelnikov, Ivan Rubachev, Artem Babenko
arXiv:2607. 16241v1 Announce Type: cross Abstract: Recent large language models (LLMs) can generate custom CUDA kernels that appear to outperform PyTorch on benchmarks such as KernelBench.
By Yunxiang Zhang (Xiangjun), Ping Yu (Xiangjun), Jianyu Wang (Xiangjun), Max (Xiangjun), Fan, Julian Reed, Azalia Mirhoseini, Will Su
TabBench-Bio is a living, interactive benchmark that evaluates machine learning models on 43 high‑dimensional biomedical tables, covering multiple domains. Using a shared cross‑validation protocol, the benchmark compares classical estimators, neural networks, and tabular foundation models across 28 feature‑by‑sample operating points, with RealTabPFN v2.5 achieving the highest performance at the reference cell of 10,000 features and 100 training samples. The benchmark provides reproducible results, fold‑level predictions, and invites community contributions to expand its dataset collection.
By Jules Kreuer, Sofiane Ouaari, Julia Hellmig, Julius Braitinger, Nico Pfeifer
AMDKernelVault is an open HIP and Triton kernel corpus and training framework designed for AMD CDNA GPUs. It includes 62,153 verified HIP kernels, 39,893 Triton kernels, and 2,377 ROCm library QA entries, and introduces agent-driven pipelines (HIPKernelGen and TritonKernelGen) that convert PyTorch references into GPU kernels, compile, validate, and profile them on AMD hardware. The corpus was used to fine‑tune Qwen3-8B, achieving the highest correctness on several benchmarks such as PyTorch-to-HIP, TritonBench‑G, and ROCmBench under fixed evaluation budgets.
By Ji Liu, Saptarshi Majumder, Yiqing Huang, Wenwen Ouyang, Umang Pandey, Zeping Li, Chushi Chen, Zihao An, Puyuan Yang, Zekai Li, Sina Rafati, Ziqiong Liu, Pratik Prabhanjan Brahma, Dong Li, Zicheng Liu, Sharon Zhou, Emad Barsoum
arXiv:2606. 04876v1 Announce Type: new Abstract: Tabular foundation models, such as TabPFN, achieve strong performance on tabular datasets with numerical and categorical data, but do not natively handle high-cardinality text features.
By Mustafa Tajjar, Alexander Pfefferle, Lennart Purucker, Frank Hutter
arXiv:2605. 18383v2 Announce Type: replace Abstract: We present TabH2O, a foundation model for tabular data that performs classification and regression in a single forward pass via in-context learning.
By Pascal Pfeiffer, Dmitry Gordeev, Mathias M\"uller, Laura Fink, Joan Salv\`a Soler, Mark Landry, Branden Murray, Marcos V. Conde, Sri Satish Ambati
The paper introduces a reproducible benchmark for evaluating attention mechanisms in tabular foundation models, focusing on the distinct row and column attention patterns that differ from language model attention. It compares several backends—Torch SDPA, FlashAttention variants, vLLM, and SageAttention—across realistic tabular shapes on A100, H100, and B200 GPUs, revealing that optimal backend choice varies by attention type, hardware, and model specifics. The study finds FlashAttention generally performs best, but CuDNN can outperform it for column attention on longer sequences, while SageAttention excels for large row sequences beyond 16k rows.
By Maximilian Schambach, Clemens Biehl, Sam Thelin
Causilo is a new tabular foundation model that delivers state‑of‑the‑art predictive performance while achieving exceptionally fast inference. On the TabArena benchmark it scores 1785.4 Elo with a median inference time of 0.10 seconds per 1 K test samples, outperforming TabPFN‑3.5‑Fast by 31.6% in speed and reaching the performance–efficiency Pareto frontier. The architecture builds on TabICL’s column‑then‑row design, adding a row‑refinement module that exchanges information among cell representations before a final column stage, and uses cross‑attention with a fixed number of summary tokens to keep attention cost linear in the number of features.
By Minyong Cho, Minho Jeong, Dooho Lee, Jinmo Lee, Jaemin Yoo
arXiv:2608. 01400v1 Announce Type: new Abstract: Tabular foundation models, driven by in-context learning, have rapidly grown in quality and popularity.
By Rasa Hosseinzadeh, Alex Labach, Zexin Xue, Shuyi Han, Valentin Thomas, Anthony L. Caterini
The paper presents Puro-2B, an open-source language model pretraining recipe that enables training models up to 1.4 trillion tokens on consumer-grade RTX 5090 GPUs using FP8 precision. The authors achieve a best model with a compute cost under $6.9K, approaching Qwen2.5-1.5B performance, and introduce a Puro Cost Scaling Law indicating that about $4.4K suffices to match Qwen2-1.5B. Additionally, they analyze how pretraining data curricula affect downstream performance, providing a full training pipeline and releasing all resources under Apache 2.0.
By Kairong Luo, Jiarui Cui, Yaorui Yin, Shengqi Chen, Yiming Yang, Linxiang Gao, Yanmohan Wang, Mingzhe Zhang, Kaiyue Wen, Kaifeng Lyu, Wenguang Chen