Causilo Technical Report

2026-09-19

AI Summary

研究通过引入Causilo模型，采用改进的列-行架构及高效推理机制，在保证高性能的同时显著降低推理时间，解决了表格数据处理中的性能与效率平衡问题。

Abstract

We introduce Causilo, a tabular foundation model (TFM) that combines frontier predictive performance with exceptionally fast inference. On TabArena, Causilo achieves 1785.4 Elo, at a median inference time of 0.10 seconds per 1K test samples. It outperforms TabPFN-3.5-Fast with 31.6% less inference time, placing it on the performance--efficiency Pareto frontier. Causilo follows TabICL's column-then-row architecture but introduces another row-refinement module before row compression. This module exchanges information among cell representations within each row after column encoding. The refined cells then visit the context set again through an additional column stage before being compressed into row embeddings. For inference efficiency, both row stages use cross-attention through a fixed number of summary tokens, keeping their attention cost linear in the number of features. Pretrained on approximately 36M synthetic tables, Causilo delivers strong benchmark results across TabArena, BeyondArena, and ScoringBench, achieving frontier-level performance with substantially faster inference.

Problem

• tabular foundation model
• inference time
• predictive performance

Innovation

• tabular foundation model
• fast inference
• row-refinement module
• cross-attention
• summary tokens