aiqbench analysis
This research introduces LUCARIO, a benchmark and Auto-BN framework that enable large language models to perform probabilistic reasoning over tabular data using neuro-symbolic methods. By automatically inducing Bayesian Networks from raw tables, the system addresses latent conditional dependencies often missed by standard NL2SQL approaches, achieving high accuracy across diverse domains. This advancement is critical for moving beyond simple factual retrieval toward handling uncertainty in structured data applications.
Towards Probabilistic Question Answering Over Tabular Data
arXiv:2506.20747v2 Announce Type: replace Abstract: Progress in question answering (QA) over tabular data has enabled reliable factual retrieval from relational tables. However, many real-world questions are probabilistic, requiring reasoning under uncertainty and latent conditional dependencies that are not directly stored in individual cells. We introduce LUCARIO, a large-scale benchmark for probabilistic QA over real-world tabular datasets, covering diverse domains, dependency patterns, and natural language variations. We propose Auto-BN, a fully automatic neuro-symbolic framework that induces a Bayesian Network from raw tables, translates natural-language questions into formal probabilistic queries, and performs exact inference. Experiments across multiple LLM backbones show that Auto-BN consistently outperforms NL2SQL, retrieval-based, and premise-based baselines, while maintaining a near-zero error rate. LUCARIO fills a relevant gap and provides a practical and scalable testbed for advancing probabilistic reasoning over structured data.
Abstract:Progress in question answering (QA) over tabular data has enabled reliable factual retrieval from relational tables. However, many real-world questions are probabilistic, requiring reasoning under uncertainty and latent conditional dependencies that are not directly stored in individual cells. We introduce LUCARIO, a large-scale benchmark for probabilistic QA over real-world tabular datasets, covering diverse domains, dependency patterns, and natural language variations. We propose Auto-BN, a fully automatic neuro-symbolic framework that induces a Bayesian Network from raw tables, translates natural-language questions into formal probabilistic queries, and performs exact inference. Experiments across multiple LLM backbones show that Auto-BN consistently outperforms NL2SQL, retrieval-based, and premise-based baselines, while maintaining a near-zero error rate. LUCARIO fills a relevant gap and provides a practical and scalable testbed for advancing probabilistic reasoning over structured data.
Accepted to AACL-IJCNLP 2026 (Main Conference). This version substantially revises and extends the original report and corresponds to the version accepted for publication. The author list has been updated to reflect the contributors to the present version
Computation and Language (cs.CL)
arXiv:2506.20747 [cs.CL]
(or arXiv:2506.20747v2 [cs.CL] for this version)
Full text under the publisher's open licence · 176 words
Related coverage
Reflections on unlearning and inoculation
LessWrong · Sep 20, 2026
Humanity’s Last Invention — Richard Socher of Recursive
Latent Space · Sep 14, 2026
Always ask your agent to be honest
LessWrong · Sep 16, 2026
Salesforce after Dreamforce: How the CRM giant can grow beyond its own interface
SiliconANGLE · Sep 19, 2026