TACT

TACT: A Complex Numerical Reasoning Benchmark

Paper - TACT: Advancing Complex Aggregative Reasoning with Information Extraction Tools

Website: https://tact-benchmark.github.io
Abstract: Large Language Models (LLMs) often do not perform well on queries that require the aggregation of information across texts. To better evaluate this setting and facilitate modeling efforts, we introduce TACT - Text And Calculations through Tables, a dataset crafted to evaluate… See the full description on the dataset page: https://huggingface.co/datasets/google/TACT.

TACT：一种复杂的数值推理基准

论文 - TACT：利用信息提取工具推进复杂聚合推理
网站：https://tact-benchmark.github.io
摘要：大语言模型（LLMs）在处理需要跨文本聚合信息查询的任务时往往表现不佳。为了更好地评估这种设置并促进建模工作，我们推出了TACT - 通过表格进行文本和计算，一个旨在评估……在数据集页面查看完整描述：https://huggingface.co/datasets/google/TACT。