Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:FLOCTKMFrecord.jsonopen to challenge →
read the original abstract
Text-to-SQL parsing, which aims at converting natural language instructions into executable SQLs, has gained increasing attention in recent years. In particular, Codex and ChatGPT have shown impressive results in this task. However, most of the prevalent benchmarks, i.e., Spider, and WikiSQL, focus on database schema with few rows of database contents leaving the gap between academic study and real-world applications. To mitigate this gap, we present Bird, a big benchmark for large-scale database grounded in text-to-SQL tasks, containing 12,751 pairs of text-to-SQL data and 95 databases with a total size of 33.4 GB, spanning 37 professional domains. Our emphasis on database values highlights the new challenges of dirty database contents, external knowledge between NL questions and database contents, and SQL efficiency, particularly in the context of massive databases. To solve these problems, text-to-SQL models must feature database value comprehension in addition to semantic parsing. The experimental results demonstrate the significance of database values in generating accurate text-to-SQLs for big databases. Furthermore, even the most effective text-to-SQL models, i.e. ChatGPT, only achieves 40.08% in execution accuracy, which is still far from the human result of 92.96%, proving that challenges still stand. Besides, we also provide an efficiency analysis to offer insights into generating text-to-efficient-SQLs that are beneficial to industries. We believe that BIRD will contribute to advancing real-world applications of text-to-SQL research. The leaderboard and source code are available: https://bird-bench.github.io/.
This paper has not been read by Pith yet.
Forward citations
Cited by 11 Pith papers
-
ACE-SQL: Adaptive Co-Optimization via Empirical Credit Assignment for Text-to-SQL
ACE-SQL jointly optimizes schema linking and SQL generation via RL with empirical credit assignment from execution-correct rollouts, achieving 65.3% greedy execution accuracy on BIRD Dev using 0.93k output tokens.
-
Data Flow Control: Data Safety Policies for AI Agents
Data Flow Control formalizes data safety as aggregate predicates over provenance monomials and implements enforcement via the Passant query rewriting layer achieving near-zero overhead across five DBMS engines.
-
Memory Architectures for Multi-Turn Text-to-SQL: A Benchmark and Empirical Study
EnterpriseMem-Bench shows stateless multi-turn Text-to-SQL accuracy drops to zero by turn 3, working memory is the main driver of gains, and additional memory components yield model- and dataset-dependent effects from...
-
CA-SQL: Complexity-Aware Inference Time Reasoning for Text-to-SQL via Exploration and Compute Budget Allocation
CA-SQL achieves 51.72% execution accuracy on the challenging tier of the BIRD benchmark using GPT-4o-mini by scaling exploration breadth according to estimated task difficulty, evolutionary prompt seeding, and candida...
-
SynQL: A Controllable and Scalable Rule-Based Framework for SQL Workload Synthesis for Performance Benchmarking
SynQL synthesizes diverse, execution-ready SQL workloads by deterministically traversing foreign-key graphs to populate ASTs, yielding high topological entropy and cost-model training data with R² ≥ 0.79 on held-out sets.
-
Spider 2.0-AIFunc: Extending Real-World Text-to-SQL to AI-Native SQL Workflows
Spider 2.0-AIFunc is a 465-instance benchmark for evaluating text-to-SQL systems on queries that incorporate Snowflake Cortex AI functions, with evaluations of ten models showing proprietary models reach 67-70% accuracy.
-
SANE Schema-aware Natural-language Evaluation of Biological Data
SANE is a new schema-aware benchmark paradigm for text-to-SQL evaluation that demonstrates few-shot LLMs with structured prompting can generate accurate queries on constrained biological data schemas without fine-tuning.
-
FlexSQL: Flexible Exploration and Execution Make Better Text-to-SQL Agents
FlexSQL reaches 65.4% on Spider2-Snow by allowing agents to flexibly explore schemas, generate diverse plans, choose SQL or Python execution, and apply two-tiered repair.
-
LeGo-Code: Can Modular Curriculum Learning Advance Complex Code Generation? Insights from Text-to-SQL
Modular curriculum learning with tier-specific adapters outperforms standard fine-tuning on complex Text-to-SQL queries in Spider and BIRD benchmarks by avoiding catastrophic forgetting.
-
MARS-SQL: A multi-agent reinforcement learning framework for Text-to-SQL
MARS-SQL trains a multi-agent RL system with ReAct-style interaction and generative validation to produce SQL queries, reaching 77.84% execution accuracy on BIRD dev and 89.75% on Spider test.
-
LLM-Based SQL Generation: Prompting, Self-Refinement, and Adaptive Weighted Majority Voting
SSEV reaches 85.5-86.4% execution accuracy on Spider benchmarks and 66.3% on BIRD-Dev through self-refinement and voting; ReCAPAgent-SQL achieves 31% on initial Spider 2.0-Lite queries via agent collaboration.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.