#database

Database

Database systems, SQL generation, and structured data management for financial AI

3 публикацииВижте всички етикети Обратно към всички публикации

8 юни 2026 г.

MAC-SQL: Multi-Agent Collaborative Text-to-SQL

MAC-SQL (COLING 2025) uses three specialized agents — Selector for schema reduction, Decomposer for question decomposition, and Refiner for execution-guided SQL correction — to reach 59.59% execution accuracy on the BIRD benchmark; ablation shows the Refiner contributes the most (+4.63 points), with direct implications for Beancount ledger query generation.

mike

aimachine-learningdatabase

7 юни 2026 г.

DIN-SQL: GPT-4 jumps 67.4%→85.3% EX on Spider

DIN-SQL lifts GPT-4 from 67.4% to 85.3% Spider execution accuracy via schema-link and self-correct stages—same decomposition fits Beancount BQL.

mike

aillmdatabase

6 юни 2026 г.

BIRD: GPT-4 hits only 54.89% EX on real databases

On BIRD, GPT-4 reaches 54.89% execution accuracy with domain hints and 34.88% without—a 20-point gap any Beancount NL→BQL interface must close.

mike

beancountaillm

Започни безплатно→Документация