Metadata Reconstruction from Values Alone: Recovering Column Semantics in Undocumented Warehouses
Mike Helwig
cs.DB, cs.AI
Submitted: 2026-08-08
Updated: 2026-08-11
Code: https://github.com/mike-m6online/Rosetta
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Reliable Text-to-SQL with Adaptive Abstention
- Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation
- Selective Classification for Deep Neural Networks
- SelectiveNet: A Deep Neural Network with an Integrated Reject Option
- On Calibration of Modern Neural Networks
- Towards Complex Text-to-SQL in Cross-Domain Database with Intermediate Representation
- Sherlock: A Deep Learning Approach to Semantic Data Type Detection
- Language Models (Mostly) Know What They Know
- RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQL
- Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs
- Calibrating LLMs for Text-to-SQL Parsing by Leveraging Sub-clause Frequencies
- RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers
- Know Your Limits: A Survey of Abstention in Large Language Models
- SQLNet: Generating Structured Queries From Natural Language Without Reinforcement Learning
- Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task
- DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correction
- What Predicts Correctness in Text-to-SQL? A Selective-Prediction Study
- Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
Related papers
- SAGA: Synthetic Agentic Graph Architecture for Temporal Benchmark Generation
- Cost-Aware Mixture-of-Experts Coordination for Model Markets
- Amora: Communication-Efficient Private Approximate Nearest Neighbors Search
- Vibe Coding on Trial: Operating Characteristics of Unanimous LLM Juries
- Human-Level Text-to-SQL via Reinforcement Learning on Verified Data, Without Pipeline Engineering
- MetaJoin: Adaptive Join Selections for Heterogeneous Tables