DataGovBench is a new benchmark using 178 large multi-tabular government datasets showing state-of-the-art LLMs and agents achieve below 40% QA accuracy and below 50% insight scores, far from real-world data analysis demands.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
V-tableR1 uses a critic VLM for dense step-level feedback and a new PGPO algorithm to shift multimodal table reasoning from pattern matching to verifiable logical steps, achieving SOTA accuracy with a 4B open-source model.
citing papers explorer
-
Data Analysis in the Wild: Benchmarking Large Language Models Against Real-World Data Complexities
DataGovBench is a new benchmark using 178 large multi-tabular government datasets showing state-of-the-art LLMs and agents achieve below 40% QA accuracy and below 50% insight scores, far from real-world data analysis demands.
-
V-tableR1: Process-Supervised Multimodal Table Reasoning with Critic-Guided Policy Optimization
V-tableR1 uses a critic VLM for dense step-level feedback and a new PGPO algorithm to shift multimodal table reasoning from pattern matching to verifiable logical steps, achieving SOTA accuracy with a 4B open-source model.