schemagate
PulseAugur coverage of schemagate — every cluster mentioning schemagate across labs, papers, and developer communities, ranked by signal.
- 2026-10-26 research_milestone Schemagate achieved the #1 ranking on the BIRD Mini-Dev leaderboard when paired with Claude Opus 5.5. source
5 day(s) with sentiment data
-
Schemagate text-to-SQL system achieves 63.7% success on Oracle databases
The Schemagate text-to-SQL system has achieved a 63.7% first-try success rate on Oracle databases, significantly outperforming previous benchmarks. This improvement is attributed to specific prompt engineering technique…
-
Developer's schemagate library outperforms other methods for text-to-SQL table selection
A developer compared four methods for selecting relevant database tables for text-to-SQL prompts, finding that their open-source library, schemagate, significantly outperformed other approaches. Using the same MiniLM mo…
-
Schemagate AI tops BIRD Mini-Dev leaderboard with Claude Opus 5.5
Schemagate, an AI system that pre-filters database tables and columns before generating SQL queries, has achieved the top position on the official BIRD Mini-Dev leaderboard. When paired with Claude Opus 5.5, Schemagate …
-
Claude Opus 5.5 achieves 77.1% on BIRD text-to-SQL benchmark with single query
A developer has achieved a 77.1% execution accuracy on the BIRD text-to-SQL benchmark using a single model call per question with Claude Opus 5.5. This approach, implemented via the schemagate system, involves the model…
-
Gemini 2.5 Flash beats Claude Sonnet 4.5 on Oracle Text-to-SQL benchmark
A new benchmark called UniQL, designed to translate natural language questions into SQL queries across various dialects, has highlighted the performance differences between large language models. On the Oracle dialect o…
-
Developer fixes LLM access control bug after benchmark reveals issues
A developer fixed a bug in their LLM access control system, schemagate, after a benchmark dataset called Spider-ACL revealed issues. The initial pilot run scored an F1 of 0.83, incorrectly refusing valid queries due to …
-
Schemagate outperforms new Text-to-SQL access control benchmark
A new benchmark for Text-to-SQL systems, released by a SIGMOD'27 paper, reveals that simply hiding restricted columns still allows models to answer questions they should refuse. The benchmark, named BIRD, showed that mo…
-
LLM reranker bug fixed, boosting text-to-SQL accuracy
A developer discovered a bug in their LLM reranker for the schemagate text-to-SQL tool where the model was being called but not providing answers due to token limits, causing the system to silently fall back to a less e…
-
Business glossary boosts text-to-SQL agent accuracy, learned terms fail
A text-to-SQL agent's performance significantly improved when provided with a comprehensive business glossary, demonstrating a 28-question gain in accuracy compared to using no glossary. However, a glossary that was aut…
-
SchemaGate 1.1.0 adds column-level access control for LLMs
SchemaGate 1.1.0 introduces column-level access control for LLMs, addressing a vulnerability where models could infer sensitive data from column names even if the table was restricted. The new feature allows users to de…